Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jun 1.
Published in final edited form as: Cell Microbiol. 2013 Dec 13;16(6):849–861. doi: 10.1111/cmi.12243

Pathogenic Rickettsia Species Acquire Vitronectin from Human Serum to Promote Resistance to Complement-mediated Killing

Sean P Riley 1,2,3, Jennifer L Patterson 1,*, Samantha Nava 1, Juan J Martinez 1,2,3,#
PMCID: PMC4028375  NIHMSID: NIHMS545750  PMID: 24286496

SUMMARY

Bacteria of the genus Rickettsia are transmitted from arthropod vectors and primarily infect cells of the mammalian endothelial system. Throughout this infectious cycle, the bacteria are exposed to the deleterious effects of serum complement. Using Rickettsia conorii, the etiologic agent of Mediterranean spotted fever (MSF), as a model rickettsial species, we have previously demonstrated that this class of pathogen interacts with human factor H to mediate partial survival in human serum. Herein, we demonstrate that R. conorii also interacts with the terminal complement complex inhibitor vitronectin (Vn). We further demonstrate that an evolutionarily conserved rickettsial antigen, Adr1/RC1281, interacts with human vitronectin and is sufficient to mediate resistance to serum killing when expressed at the outer-membrane of serum sensitive E. coli. Adr1 is an integral outer-membrane protein whose structure is predicted to contain eight membrane-embedded β-strands and four “loop” regions that are exposed to extracellular milieu. Site-directed mutagenesis of Adr1 revealed that at least two predicted “loop” regions are required to mediate resistance to complement-mediated killing and vitronectin acquisition. These results demonstrate that rickettsial species have evolved multiple mechanisms to evade complement deposition and that evasion of killing in serum is an evolutionarily conserved virulence attribute for this genus of obligate intracellular pathogens.

Keywords: Rickettsia, complement, vitronectin

INTRODUCTION

Members of the spotted fever group (SFG) Rickettsia are small gram-negative, obligate intracellular α-protebacteria that are typically transmitted to mammalian hosts via an arthropod vector. The “spotted fever” nomenclature pertains to the characteristic maculopapular dermal rash frequently associated with disseminated disease. This rash is a physical indicator of more severe underlying pathology linked to infection of the endothelial lining, dissemination throughout many organs, and subsequent inflammatory processes (Walker et al., 1987). Severe manifestations of vascular discontinuity induced by Rickettsia infection include renal failure, pulmonary edema, interstitial pneumonia, and other multi-organ manifestations (Chapman et al., 2006). While diagnosis of SFG rickettsial infection is frequently successful if spotted fever or eschar is observed, misdiagnosis is frequent and is associated with poor medical outcome (Helmick et al., 1984). Misdiagnosis and inappropriate treatment results in mortality rates reported to be as high at 32% (Dalton et al., 1995).

As obligate intracellular pathogens, SFG rickettsial species proliferate within the cytosol of the amplifying mammalian host or arthropod vector. While these species require the intracellular environment for growth, they are thought to spend periods outside the relative protection of the host cytoplasm. During these periods outside of a host cell, the bacteria are frequently immersed in blood and are therefore subjected to the anti-bacterial effects of the mammalian complement cascade.

Host complement is a key component of the innate immune system that includes both anti-and pro-inflammatory properties (Walport, 2001b; Walport, 2001a). Complement includes a direct microbial killing mechanism that is lethal in the absence of microbial countermeasures (Lambris et al., 2008; Zipfel et al., 2009). Additionally, complement activation also results in production of major anaphylatoxins and opsinophagocytosis of the detected pathogen (Ember et al., 1997). The initial steps of complement consist of three independent activation cascades. Each of these cascades converges at C3 deposition on a target surface and its conversion to an unstable protease called C3 convertase. This protease initiates the proteolytic cascade for deposition of antimicrobial pore-like structure deemed the terminal complement complex (TCC) (Tegla et al., 2011).

We have previously demonstrated that R. conorii is resistant to normal serum complement (Chan et al., 2011). This bacterial phenotype resembles the protected state of host cells, which possess anti-complement effectors (Meri et al., 1998). Indeed, some bacteria utilize these host serum regulatory proteins for their own benefit (Blom et al., 2009; Lambris et al., 2008). We have previously ascertained that the β-peptide or translocon domain of the R. conorii autotransporter protein OmpB mediates acquisition of a complement regulatory protein, factor H, and this interaction is sufficient to mediate resistance to the bactericidal effects of complement (Riley et al., 2012). While depletion of factor H from serum provided some loss of serum resistance of this pathogen, R. conorii still remained significantly resistant to serum challenge. This phenotype indicates that R. conorii possesses other mechanisms to evade complement-mediated elimination from the host pulmonary circulation.

RESULTS

Adr1 conservation and predicted structure

The protein encoded by the R. conorii open reading frame RC1281 was previously demonstrated to interact with an unknown mammalian protein and subsequently named Adr1 (Renesto et al., 2006). Interestingly, analysis of the Adr1 secondary structure indicated structural but insignificant amino acid similarity to a family of gram-negative outer membrane proteins embodied by E. coli OmpX (Vogt et al., 1999). This protein family consists of integral outer membrane proteins with 8 membrane-spanning β-sheets which form a barrel-like structure reminiscent of autotransporter β-peptides. Surface exposed peptide loops connect the transmembrane β-sheets (Jacob-Dubuisson et al., 2004). Some members of this protein family, namely Yersinia enterolitica Ail and Salmonella typhimurium Rck, are sufficient to confer serum resistance through acquisition of host complement regulatory proteins (Bartra et al., 2008; Ho et al., 2010). Based on the observed structural similarity to other proteins, we hypothesized that Adr1 is also expressed at the rickettsial outer membrane and is capable of interacting with a serum regulatory protein.

Initial analysis of Adr1 primary structure from various Rickettsia species indicated significant amino acid identity across the genus Rickettsia (Figure 1A). When comparing the R. conorii Adr1 (RC1281) amino acid sequence to homologs from R. rickettsii (RR7045), R. typhi (RT815), R. prowazekii (RP827) we observe 96.0%, 69.3%, 69.5% identity and 97.6%, 80.5%, 79.9% similarity, respectively. Each of these proteins retains near complete identity in the predicted transmembrane β-sheets, and large stretches of identity in the interconnecting loops. When these primary and secondary structures are applied to a Phyre-constructed model of Adr1 tertiary structure (Figure 1B), we clearly observe the “barrel”-like transmembrane regions indicated in yellow and surface-exposed loops in red. The barrel-like structure (yellow) exhibits a hydrophobic outer surface that likely serves to interact with outer membrane lipids.

Figure 1. Adr1 is conserved in pathogenic rickettsiae.

Figure 1

Open in a new tab

A. Amino acid alignment of Adr homologs from various Rickettsia species from both the spotted fever group (R. conorii, R. rickettsii) and typhus group (R. typhi, R. prowazekii) demonstrates significant conservation throughout these proteins. Indicated above the sequences are the predicted secondary structures, including transmembrane β-sheets (yellow arrows) and peptide loops extending outside of the outer membrane (red brackets). B. The predicted structure of R. conorii Adr1 demonstrates a “barrel”-like structure with 8 transmembrane β-sheets (yellow) and likely surface-exposed peptide loops (red).

Adr1/RC1281 is present at the R. conorii surface

In order to query for the presence of Adr1 in Rickettsia species, we produced polyclonal antiserum directed against small peptides of the Adr1 protein that are both conserved in rickettsial species and are predicted as extracellular domains. As shown in Figure 2A, immunoblot interrogation of rickettsial cell lysates yields an anti-Adr1 reactive band present at the predicted molecular weight (27kDa) in both R. conorii Malish 7 and R. rickettsii Sheila Smith. To verify surface exposure of Adr1, we probed paraformaldehyde-fixed R. conorii with the above-described anti-Adr1 antibody and fluorescently labeled secondary antibody. Through flow cytometric analysis (Figure 2B), we were able to demonstrate the presence of Adr1 at the surface of R. conorii (green trace), while control untreated bacteria (red trace) or R. conorii treated only with secondary antibody (blue trace) did not exhibit significant reactivity. Furthermore, we confirmed the presence of Adr1 in the R. conorii outer membrane by western immunoblot analysis of isolated total outer membrane fractions (Figure 2C). Taken together, these results indicate that Adr1 is present in the rickettsial outer membrane and that portions this protein are exposed to the extracellular milieu.

Figure 2. Outer membrane (OM) expression of Adr1 in R. conorii.

Figure 2

Open in a new tab

A. Western immunoblot analysis using anti-Adr1 antiserum confirms expression in R. conorii Malish 7 and R. rickettsii Sheila Smith. B. Flow cytometric analysis confirms expression of Adr1 at the surface of R. conorii. A shift in fluorescence intensity was only observed when both primary and secondary antibodies were applied to the bacteria (green) as opposed to samples lacking any antibody (blue) or incubated with secondary antibody alone (red). C. Western immunoblotting on R. conorii whole cell lysates (WCL) and OM fractions using antibodies against OmpA (mAb 13.3), OmpB (mAb 6B.6), and Adr1 revealed the presence of reactive species in both the WCL and OM fractions. Anti-RlpF (50s ribosomal protein L6) was used as a control for cytoplasmic contents.

Adr1 mediates serum resistance

We next sought to determine if Adr1 expression in a serum sensitive strain of E. coli is sufficient to confer serum resistance. We cloned the R. conorii adr1/RC1281 open reading frame into the plasmid pET22b to form pJP01, which encodes for an Adr1 protein fusion consisting of a N-terminal E. coli-optimized secretion signal (PelB) and C-terminal His6 tag. Upon expression of this adr1 construct in E. coli BL21, anti-His6 reactive species of the appropriate molecular weight appear in the outer membrane (OM) protein fractions (Figure 3A). The OM fractions derived from E. coli are free of cytoplasmic contents as demonstrated by presence of the cytoplasmic protein RNA polymerase α-subunit (RNAP) exclusively in the whole cell lysate control lane. To verify surface exposure of Adr1 when expressed in E. coli, we incubated these bacteria with anti-Adr1 and appropriate fluorescently-labeled secondary antibody. As shown in Figure 3B, flow cytometric analysis indicates an increase in fluorescence associated only with Adr1-expressing bacteria (pJP01, blue trace) and not with those bacteria containing the empty vector pET22b (red trace). Together, these results demonstrate that E. coli expressed Adr1 localizes to the bacterial outer membrane and contains surface exposed epitopes.

Figure 3. Expression and serum resistance of Adr1 in E. coli.

Figure 3

Open in a new tab

A. Outer membrane (OM) preparations or whole cell lysate (WCL) from E. coli harboring the empty vector (pET22b) or plasmid encoding for His6-tagged Adr1 (pJP01). Coomassie staining of the SDS-PAGE demonstrates similar protein loading (upper panel). Anti-E. coli RNA Polymerase alpha (RNAP) is used to demonstrate the presence or absence of bacterial cytoplasmic contents (middle panel). The associated anti-His6 western blot yields reactive bands at the appropriate molecular weight for Adr1 expression that is present only in the RNAPα-free pJP01 OM preparations (bottom panel). B. Flow cytometry reveals surface expression of Adr1 on E. coli BL21 harboring pJP01 (Blue) but not the empty vector pET22b (Red). C. Expression of Adr1 in BL21 is sufficient to mediate resistance to normal human serum (NHS).

Finally, to determine if presence of Adr1 in the E. coli outer membrane confers serum resistance, we incubated Adr1 expressing E. coli with normal human serum for one hour at 37°C with aggressive shaking. Viable bacteria remaining after serum challenge were quantified by plating, and values are expressed as % survival after serum challenge. As seen in Figure 3C, bacteria harboring pJP01 are resistant to serum killing when compared to bacteria containing the empty vector (pET22b). These results demonstrate that Adr1 is sufficient to mediate serum resistance when expressed in E. coli.

R. conorii acquisition of soluble complement regulators

We have previously demonstrated that R. conorii is intrinsically resistant to the deleterious effects of host complement and that this bacterium acquires the complement regulator factor H (Chan et al.,2011; Riley et al., 2012). However, depletion of the protective factor H protein from human serum only resulted in a partial increase in R. conorii sensitivity to serum complement (Riley et al., 2012). As such, we hypothesized that R. conorii attains additional host complement regulatory factors. To assess this hypothesis, we incubated R. conorii with normal human serum (NHS), and queried for the co-sedimentation of complement regulatory proteins with the intact bacteria. By immunoblot, we observed the two characteristic vitronectin (Vn) SDS-isoforms of the appropriate molecular weight in samples containing the positive control (NHS alone) and R. conorii incubated with NHS (Figure 4A). Since these bands were not present in the R. conorii incubated with PBS, we can conclude that R. conorii acquire Vn from serum. Additionally, as a control, we analyzed NHS treated R. conorii lysates for acquisition C1 Esterase Inhibitor (C1EI), but did not observe deposition of this host complement-regulatory protein (Figure 4A).

Figure 4. R. conorii acquisition of soluble complement regulatory proteins.

Figure 4

Open in a new tab

A. Western immunoblot analysis reveals that vitronectin interacts with R. conorii. NHS is used as a control for the antibodies used alone. As a control, C1 Esterase inhibitor (C1EI) is only observed in the NHS lane, indicating that, by this metric, R. conorii does not acquire this complement-regulatory protein. B. Flow cytometry confirms the interaction of vitronectin with R. conorii as revealed by an increase in fluorescence intensity of the NHS-treated (blue) over the PBS-treated (red) R. conorii. 59.9% of NHS-treated and 15.7% of PBS-treated bacteria are included in the bracketed area.

We confirmed the deposition of Vn on the R. conorii surface using flow cytometry. Bacteria were incubated with NHS and processed for flow cytometric analysis using anti-Vn antibody and fluorescently labeled secondary antibody. As shown in Figure 4B, we observed increased fluorescence of the NHS-treated samples (blue trace) over PBS-treated counterparts (red trace), thereby demonstration of Vn deposition on the surface of R. conorii.

Expression of Adr1 in E. coli promotes vitronectin binding

To query the potential connection between R. conorii vitronectin (Vn) acquisition and Adr1-mediated serum resistance, we asked if expression of Adr1 in the E. coli OM is sufficient to mediate acquired interaction with Vn. E. coli harboring pET22b or pJP01 were incubated with ice-cold NHS. The use of cold NHS permits potential protein-protein interactions while blocking the proteolytic cascade, which would result in the ablation of Adr1-free bacteria. Total cellular lysates from these serum-incubated bacteria were prepared, separated by SDS-PAGE, and the presence of Vn was queried by western blot. As shown in Figure 5A, incubation with anti-Vn and associated secondary antibody yields the two characteristic Vn bands in the positive control, NHS (right) (Peterson, 1998). Adr1-expressing bacteria demonstrated Vn reactivity, while empty vector (pET22b) bacteria failed to interact with Vn. We additionally probed for the presence of another complement regulatory proteins C1 Esterase Inhibitor (C1EI), but as with R. conorii, did not observe Adr1-mediated acquisition of this protein.

Figure 5. Adr1 expression in E. coli promotes Vitronectin binding.

Figure 5

Open in a new tab

A. Western immunoblot analysis reveals that E. coli BL21 harboring pJP01, but not pET22b interacts with vitronectin. Adr1 expression is demonstrated through anti-Adr1 and anti-His6 blots. Quantity of E. coli samples is indicated by the presence of RNAPα. Another complement-regulatory protein, C1 Esterase inhibitor (C1EI) does not interact with Adr1-expressing bacteria. B. Flow cytometry analysis of BL21(pET22b) (Red, 2.67% gated) and BL21(pJP01)−IPTG (Blue, 5.76% gated) showed no difference in vitronectin deposition. However, induced BL21(pJP01)+IPTG (green, 13.2% gated) showed a large increase in fluorescence intensity, demonstrating vitronectin acquisition by these bacteria.

To confirm Vn association with Adr1-expressing E. coli, we again incubated E. coli with ice-cold NHS and the bacteria processed for analysis by flow cytometry. As shown in Figure 5B, Adr1-expressing E. coli (green trace) bacteria exhibited greater fluorescent signal than the uninduced (−IPTG, blue trace) or empty vector (pET22b, red trace) bacteria. Taken together, these results demonstrate that production of Adr1 promotes serum resistance and that this activity is correlated with the ability of Adr1 to interact with Vn.

Additionally, we queried the ability of the R. rickettsii (RR7045), R. typhi (RT815), and R. prowazekii (RP827) Adr1 homologs to mediate Vn acquisition and serum resistance when expressed in E. coli. While the R. conorii and R. rickettsii Adr1 homologs are nearly identical, the R. typhi and R. prowazekii homologs demonstrate less identity and ~80% amino acid similarity. As anticipated, R. rickettsii Adr1 mediated Vn acquisition and serum resistance to levels similar to that found with R. conorii Adr1 (Supplemental Figure 1). In addition, the R. typhi and R. prowazekii Adr1 proteins also promoted Vn acquisition and serum resistance (Supplemental Figures 2 and 3). These findings indicate that vitronectin binding and serum resistance are conserved attributes of Adr1 homologues expressed by pathogenic rickettsial species.

Mapping Adr1 elements required for serum resistance and vitronectin acquisition

We next sought to establish the regions in Adr1 that contribute to serum resistance. To this end, we utilized a genetic approach to remove the predicted extracellular peptide loop regions in Adr1 (red, Figure 1B) that are likely to interact with the extracellular milieu while leaving long enough amino acid stretches to permit the required connection between transmembrane β-strands. Through successive deletion of each of the extracellular loops, we created pJP01 derivatives that retain a single loop or no loops at all (Figure 6A). The pJP01-L# nomenclature refers to the intact loops remaining whereby pJP01-L1/2 retains loop 1 and 2, pJP01-L2 retains loop 2, etc. pJP01-L0 does not encode for any intact extracellular loops. The pJP01-L1 plasmid was constructed, but the encoded protein was not produced as efficiently as the other Adr1 derivatives, and was not further analyzed.

Figure 6. Identification of Adr1 extracellular loops responsible for serum resistance and vitronectin acquisition.

Figure 6

Open in a new tab

A. The pJP01 derivative plasmid nomenclature indicates loop regions that remain intact in each plasmid construct. The connecting lines indicate loop regions that are missing from each plasmid construct with deleted amino acids indicated at the top. B. Outer membrane (OM) preparations or whole cell lysate (WCL) from E. coli harboring the empty vector (pET22b) or plasmid encoding for His6-tagged Adr1 (pJP01-L#) derivatives. Anti-E. coli RNA Polymerase alpha (RNAP) is used to demonstrate the presence or absence of bacterial cytoplasmic contents in each control WCL lane. Anti-His western blot analysis confirms the expression of constructs at the E. coli OM. C. Serum survival assays demonstrate that Adr1 loops 3 and 4, but not 1 or 2 are sufficient to mediate resistance to killing in normal Human serum (NHS) when expressed in E. coli. pJP01-L3 and pJP01-L4=12% survival, pJP01-L1/2=0.012%, pJP01-L1=0.00012%, pJP01-L2=0.0015%, pJP01-L0=0.00017%. Statistical analysis by 1 way ANOVA with Dunnett’s multiple comparison indicated that only pJP01-L3 and pJP01-L4 samples are statically different than pET22b. D. Vitronectin binding assays indicate that E. coli expressing Adr1 or variants containing loop 3 or 4 (pJP01-L3/pJP01-L4) are able to sequester vitronectin from NHS. Conversely, strains lacking Adr1 (pET22b) or possessing only loops 1 and 2 (pJP01-L1/2) associate poorly with vitronectin. RNA polymerase (RNAP) signal indicates total E. coli in each lane and is utilized as a loading control. The lower panel illustrates densitometric analysis of the western blot data whereby the ratio of vitronectin and control RNAP signals are graphed for each Adr1 derivative.

We initially verified that each construct was present at the OM when expressed in E. coli. As shown in Figure 6B, we were able to confirm the presence of Adr1 derivatives at the OM of E. coli through western blotting with anti-His6. Again, verification that the outer membrane extractions are free of cytoplasmic contents was demonstrated by the presence of the cytoplasmic protein, RNA polymerase α-chain (RNAP) solely in the whole cell lysate control lanes.

To interrogate the ability of each of the Adr1 derivatives to mediate serum resistance, Adr1-expressing E. coli were incubated in NHS. Input and post-serum viable bacteria were quantified and all values are expressed as percent bacteria remaining after serum challenge. Bacteria lacking any Adr1 (pET22b) are sensitive to serum complement as demonstrated by over 6 log10 killing (1.78E-05% survival). In contrast, E. coli harboring either pJP01-L3 or pJP01-L4 are resistant to serum with approximately 12% viable bacteria remaining after serum challenge, while the other Adr1 derivatives remain sensitive to serum killing. Taken together, we can conclude that either loop3 or loop4 is sufficient to mediate serum resistance in the overall context of outer membrane localized Adr1. Conversely, Adr1 derivatives lacking all extracellular loops or containing any combination of loops 1 or 2 are insufficient to promote serum resistance.

Subsequently, we hypothesized that serum resistance phenotypes would correlate with Adr1-mediated vitronectin (Vn) acquisition. As described in Figure 6D, we again utilized cold NHS to permit Vn interaction while preserving complement sensitive strains. By this metric, E. coli expressing Adr1 or derivatives containing extracellular loop 3 or loop 4 were able to sequester vitronectin. Conversely, bacteria lacking Adr1 or expressing an Adr1 derivative containing only loop 1 and 2 were unable to interact with Vn. As such, we conclude that Adr1-derivative serum resistance phenotypes correlate with Vn acquisition.

DISCUSSION

Our results have demonstrated that R. conorii Adr1 serves to mediate resistance to host complement in a mechanism that correlates with acquisition of the complement-regulatory protein vitronectin (Vn). Adr1 is structurally related to a family of integral outer membrane proteins that form barrel-like structures with 8 membrane-spanning β-sheets and interconnecting loops. Four of these peptide loop regions are predicted to reside outside of the bacterial outer membrane and therefore have the potential to interact with molecules of the extracellular milieu. Our first results confirmed expression, surface-exposure, and outer membrane localization of Adr1 in spotted fever group Rickettsia species. We subsequently hypothesized that Adr1 acts to promote serum resistance in a mechanism similar to other bacterial integral outer membrane proteins. This hypothesis led us to heterologous expression of Adr1 in serum sensitive E. coli. After validation of Adr1 outer membrane localization and surface exposure, these E. coli were challenged with normal human serum. Expression of Adr1 in E. coli conferred serum resistance phenotype upon this serum sensitive bacterium.

To gain insight into the potential mechanism of Adr1-mediated serum resistance and to survey the ability of R. conorii to acquire complement regulatory proteins, we queried R. conorii for the ability to associate with complement regulatory proteins found in normal human serum (NHS). Of the host proteins tested, the terminal complement complex inhibitor vitronectin (Vn) deposited on R. conorii as evidenced through western blotting and flow cytometry. Correlatively, expression of Adr1 in E. coli provided these bacteria with the ability to sequester Vn from human serum. Additionally, Adr1 homologs from various Rickettsia species also endowed Vn acquisition and serum resistance phenotypes upon E. coli. As such, we conclude that the Rickettsia Adr1 proteins mediate resistance to human serum in a mechanism that includes vitronectin acquisition.

Vitronectin (Vn) is a multifunctional mammalian protein with roles in cell anchoring, hemostasis, coagulation, peri-cellular proteolysis, leukocyte recruitment, and complement regulation (reviewed in (Preissner et al., 2011)). In spite of these multiple functions, the most apparent result of an interaction between any bacteria and host Vn while immersed in serum is inhibition of the terminal complement complex (TCC). In fact, some bacteria have documented ability to sequester this molecule from serum to promote serum resistance (Singh et al., 2010).

Previously, we demonstrated that R. conorii is inherently resistant to complement-mediated killing in a mechanism that includes bacterial acquisition of the complement regulatory protein factor H (Chan et al., 2011; Riley et al., 2012). However, inhibition of factor H acquisition did not make R. conorii as sensitive to serum challenge as most other bacterial species. We therefore hypothesized that R. conorii remains partially resistant to serum complement through acquisition of other host complement regulatory proteins. We now propose that both R. conorii OmpB β-peptide and Adr1 contribute to serum resistance of this bacterium by binding factor H and vitronectin, respectively. While both of these host proteins act to interrupt the complement cascade, each works at a different step in the process. Factor H functions early in the complement cascade through disruption of C3 convertase activity (Pangburn et al., 1977; Weiler et al., 1976). In contrast, vitronectin inhibits later in the complement cascade at the pore-forming terminal complement complex (C5b-9) (Preissner et al., 2011). Therefore, R. conorii inhibits the complement activation at two stages of the cascade and both molecular interactions promote overall rickettsial serum resistance. Extensive examination will be conducted in order to determine the relative contribution of each of these protein-protein interactions in successful achievement of the R. conorii lifecycle.

As seen in the Adr1 model in Figure 1B, the protein appears to form a barrel-like structure. The extracellular loops protrude far away from the outer leaflet of the outer membrane with peptides consisting of 18–24 amino acids. We interpret the existence of these larger extracellular loops as the Adr1 regions with the most potential for interaction with the extracellular milieu. Through successive mutagenesis we produced Adr1 proteins retaining loops 1 and 2, loop 2, 3, 4, or no extracellular loops. Interestingly, loop 3 or loop 4 could independently promote serum resistance in the overall context of the Adr1 backbone when expressed in E. coli, while Adr1 derivatives containing loop 1 and 2, loop 2 alone, or no loops demonstrated serum sensitivity similar to Adr1-free E. coli. Of note, the extracellular loops 3 and 4 of Adr1 are positively charged with an over-representation of Lysine residues. The ~24 amino acid loop 3 has a predicted pI of 9.70 and contains 25% lysines while the ~25 a.a loop 4 has a predicted pI of 9.88 and contains 20% lysine residues. This general positive charge of the extracellular loops likely provides the molecular interface for interaction with vitronectin. Expression of Adr1 derivatives containing only extracellular loop 3 or loop 4 were also sufficient to mediate interaction with vitronectin as compared to Adr1 derivatives containing loops 1 and 2. We can therefore conclude that Adr1 mediated serum resistance correlates with vitronectin acquisition.

An important aspect of this work pertains to the relationship between obligate intracellular rickettsial species and the extracellular environment. By definition, these bacteria reside inside the host cytoplasm to perform many essential bacterial processes. As such, the limited examination of extracellular Rickettsia has focused only on assessing the bacterial factors required for invading a host cell (Chan et al., 2010). While the quantity of time spent outside the host cell in an in vivo setting has not been assessed, there are periods in the Rickettsia lifecycle where the bacteria reside outside of the host cell. For example, a Rickettsia-infected animal would likely experience some period of bacteremia to facilitate the infection of a feeding arthropod. In addition, during the periods of transmission from a feeding rickettsiae-infected arthropod to the mammalian host, the bacteria are likely immersed in blood (Raoult et al., 1997) and are likely exposed to the deleterious effects of complement deposition. Rickettsiae invade host cells by induced phagocytosis (Walker et al., 1978), a process that involves extracellular bacteria, presumably free in circulation, invading host endothelial cells and other target host cells. Interestingly, typhus group Rickettsia multiply within and eventually lyse infected host cells for subsequent spread to other naïve cells, thereby potentially exposing themselves to the bactericidal effects of the complement system (Bechah et al., 1998). While these bacteria do spend an indeterminate time outside of the relative protection of the host cytoplasm, the extent, location, and function of extracellular Rickettsia in an in vivo setting currently suffers from a lack of understanding and will be a future topic of investigation.

In this work, we have examined vitronectin (Vn) in relation to its complement inhibitory properties. However, Vn contributes to many other host processes. For example, Vn-coated R. conorii would have the potential to interact with a plethora of other host proteins, for which the potential phenotypic outcomes are diverse. Vn has been demonstrated to interact with the mammalian cell surface molecules heparin, alphav intergins, and urokinase receptor (reviewed in (Preissner et al., 2011)). As such, a potential function of R. conorii-bound Vn association is bacterial adherence to host cells. Adr1 has been identified through proteomic analysis to associate with an unknown factor found in biotinylated mammalian cell extracts (Renesto et al., 2006) and binding of Vn has been associated with host cell adherence in some other gram-negative bacteria, including H. influenzae, M. catarrhalis, N. gonnorhoeae, and other species (reviewed in (Singh et al., 2010)). While multiple R. conorii adhesins have previously been identified (Chan et al., 2010), the Adr1-Vn interaction could potentially aid in this essential bacterial function.

Another interesting potential function associated with the R. conorii-Vn interaction relates to the propensity for rickettsial infections to increase microvascular permeability, resulting in the most severe manifestations of disease (Walker et al., 1987). Recently, vitronectin has been implicated in elevation of vascular permeability through interaction with alphav integrins and subsequent host cell signaling (Li et al., 2012; Tsuruta et al., 2007). It will be intriguing to explore the relationship between rickettsial acquisition of vitronectin and induced vascular permeability.

One exciting extrapolation of the data presented in this study pertains to the potential of Adr1 as a target for vaccination. Adr1 loops 3 and 4 are well conserved throughout “spotted fever group” Rickettsiae. Overall, Adr1 exhibits 96, 94.8, 92.0, 96.0, 97.6, and 94.1% identity with its homologs from R. rickettsii Sheila Smith, R. africae ESF, R. massiliae MTU5, R. slovaca 13-B, R. parkeri Portsmouth, and R. japonica YH, respectively. Analysis of loop 3 in these species demonstrates that all sequences are identical except for R. japonica, which has a single difference at R. conorii Adr1 residue 177. Loop4 is similarly conserved with the exception of a mix of asparagine (N) or glycine (G) residues present at R. conorii Adr1 position 231. Taken together, these results suggest that utilizing peptides derived from Adr1 loops 3 and 4 may be an efficacious strategy to generate broadly protective immune responses against fatal rickettsial infections.

In summary, we have identified a valuable function for a highly conserved rickettsial protein. R. conorii Adr1 is capable of mediating serum resistance in a mechanism that includes binding of the host complement inhibitory protein vitronectin. Through mutagenesis we have additionally determined the surface exposed peptides that are sufficient to mediate serum resistance and vitronectin acquisition. Through analysis of multiple Adr1 proteins, we can posit that serum resistance and vitronectin binding are universal functions of highly pathogenic Rickettsia found throughout the world.

EXPERIMENTAL PROCEDURES

Protein alignment and modeling

Alignments from Rickettsia spp. were performed with MacVector (MacVector, Inc) using the input accession sequences (R. africae ESF-5 YP_002845692.1, R. parkeri Portsmouth YP_005393487.1, R. sibirica 246 ZP_00142605.1, R. slovaca 13-B YP_005066302.1, R. rickettsii Sheila Smith YP_001495358.1, R. japonica YH YP_004885269.1, R. prowazekii Madrid E NP_221176, R. typhi Wilmington YP_067752) (Fournier et al., 2009). The Adr1 peptide sequence consisting of R. conorii amino acids 42–248 was processed by the Phyre server for the purpose of creating tertiary structure models (Kelley et al., 2009). Secondary structure predictions were estimated using the Phyre output. The algorithm yielded Adr1 models based on crystal structure of Neisserial NspA (pdb1p4t) (Vandeputte-Rutten et al., 2003), E. coli OmpA (pdb1bxw) (Pautsch et al., 1998), E. coli OmpW (pdb2ft1t) (Hong et al., 2006). Likely membrane localization and surface exposure was determined by modeling the hydrophobic interfaces of the transmembrane β-strands based on Jmol analysis (Jmol: an open-source Java viewer for chemical structures in 3D. http://www.jmol.org/).

Antibodies

Anti-Adr1 peptide polyclonal antiserum was generated by immunizing rabbits with peptides CKIHSKDIKGGVTDTNFGTTKNKTN and CSWRDYGKTKNTTKTINGDK, corresponding to Adr1 amino acids 163–186 (loop3) and 210–228 (loop 4)(Yenzym Antibodies). Total antibodies were recovered from the immune serum utilizing HiTrapProteinA columns (GE healthcare). Polyclonal antiserum raised against the R. conorii Malish 7 50s ribosomal protein L6 (RplF) was generated in rabbits using the peptide KQRPPEPYKGKGIKFENQ corresponding to amino acids 150–167 in the R. conorii RplF sequence (NP_360628.1). Monoclonal antibodies against rOmpB (mAb 6B.6) and rOmpA (mAb 13-3) have been described (Anacker et al., 1987; Chan et al., 2011).

Plasmid construction

The adr1 (RC1281) open reading frame was amplified from R. conorii Malish 7 genomic DNA using the primers RC1281F and RC1281R (Supplemental Table 1) and cloned into pCR2.1 resulting in plasmid pCR2.1-RC1281. The adr1 DNA fragment was liberated from pCR2.1 using BamHI and NcoI, and ligated into a similarly digested pET22b to make plasmid pJP01. pRR7045 was constructed similarly using RC1281F and RR7045R. pRP827 was constructed using SRK-RP827F and SRK-RR7045R. pRT815 was constructed using SRK-RT815F and SRK-RT815R. The encoded proteins contain N-terminal E. coli pelB signal sequence, adr1 homolog lacking a rickettsial signal sequence, and C-terminal His6 tag under the control of the IPTG-inducible T7promoter.

pJP01 was used as a template for quick-change site-directed mutagenesis to remove the DNA encoding for the extracellular loops of Adr1 while leaving sufficient amino acids to allow small extracellular turns, thereby preserving overall tertiary structure. Deletion primers are noted in Supplemental Table 1, whereby dL# nomenclature indicates the primers used to delete each loop sequence i.e. 1281-dL1-1 and RC1281-dL1-2 were used to delete loop 1. In all cases, the PCR products were digested with DpnI to remove the parent plasmid, and transformed into MaxEfficiency DH5α-T1R (Life Technologies). Loops were sequentially deleted in this fashion until plasmids encoding for Adr1 with single loop or no loops were created. pJP01 derivatives are named for the loop(s) remaining. Amino acid deletions are designated in Figure 6A.

Bacterial growth

R. conorii Malish7 and R. rickettsii Sheila Smith were propagated from Vero cells and purified as described previously (Ammerman et al., 2008, Chan et al., 2009, Chan et al.). The yielded bacteria were pure and free of host contamination as visualized by microscopy. E. coli BL21(DE3)(pJP01, pET22b, or pJP01-L# derivatives) were grown overnight in LB+ 50μg/mL carbenicillin 37°C. The bacteria were diluted 1:10 in fresh media, grown to OD600 ~0.5, and induced with 0.5mM IPTG for 3.5 hours at 37°C where appropriate.

Bacterial Fractionation

Briefly, approximately 5×106 plaque forming units (pfu) of purified R. conorii Malish 7 was fixed in 4% paraformaldehyde (PFA) in PBS, washed in PBS and then removed from BSL3 containment after verification that viable rickettsiae were no longer present according to standard operating procedures (SOPs). For whole-cell lysates, these bacteria were resuspended in SDS-PAGE buffer and boiled. Total outer membrane proteins were extracted essentially as described in (Nikaido, 1994). The sample was resuspended in 1.5ml of 20mM Tris pH 8.0 containing 1X protease inhibitor cocktail and then subjected to three rounds of French pressure cell lysis. The resulting lysate was centrifuged at 10,000 g for 3 minutes to remove unbroken cells and then incubated in 0.5% sarkosyl at room temperature for 5 minutes. The sarkosyl-treated lysate was centrifuged at >16,000 g for 30 minutes at 4°C. The sarkosyl soluble protein fraction was removed and the remaining insoluble pellet representing the outer-membrane protein fraction was washed in 20mM Tris pH 8.0 and then boiled in 0.5ml of 20mM Tris pH 8.0 containing 0.5ml of 2X SDS sample buffer. Protein samples were aliquoted and frozen at −20°C until use. Outer-membrane (OM) protein fractions E. coil were generated essentially as described as above (Nikaido, 1994; Thanassi et al., 1998).

OM proteins were separated by SDS-PAGE, transferred to nitrocellulose, and immunoblotted with polyclonal rabbit anti-His6 antibody (PRB-156C, Covance) and goat anti-rabbit IgG HRP. Removal of E. coli cytoplasmic contents was verified through western blot using antibodies against the cytoplasmic E. coli RNA polymerase alpha subunit (4RA2, Neoclone, Madison, WI). RplF antibody was utilized to verify the presence of R. conorii cytosolic contents.

Flow Cytometric Detection of Adr1

Paraformaldehyde-fixed R. conorii or E. coli harboring pJP01 were used to probe for Adr1 surface-exposure. Adr1 was detected using anti-Adr1, rabbit anti-goat-Alexafluor488 (Molecular Probes), and 5μg/mL DAPI. The samples were analyzed with a LSR-II cytometer (BD Biosciences, Franklin Lakes, NJ) using FITC (488nm excitation 530/30nm emission) and DAPI (325nm excitation 450/50nM emission) fluorescence settings. All events displayed were positive for DAPI staining. All samples were analyzed with FloJo software (Tree Star, Ashland, OR).

Detection of host complement regulatory protein deposition

R. conorii were incubated in PBS or a 50% NHS/PBS mixture for 1 hour at room temperature with shaking. E. coli were similarly treated, except for the NHS was kept ice-cold to prevent complement deposition and subsequent bacterial killing. Each sample was washed twice and then lysed by boiling in SDS-PAGE sample buffer. Total cellular proteins were separated by SDS-PAGE, transferred to nitrocellulose and then immunoblotted with antibodies against vitronectin or C1 Esterase Inhibitor (Complement Tech). Secondary antibodies for immunoblot included Donkey-anti-Rabbit-HRP, Donkey anti- Mouse-HRP, Donkey anti-Goat-HRP (Sigma), IRDye 800 CW Donkey anti-Goat IgG/IR Dye 800CW Donkey anti-Rabbit IgG and were analyzed using Odessey CLx (LiCor) instrument or exposure to film. Expression of Rickettsia sp. Adr1 was confirmed using anti-His Antibody (Covance or Genscript) or above described anti-Adr1 antibody.

For flow cytometric analysis of vitronectin deposition, R. conorii were incubated in PBS or 50%NHS for one hour at 37°C with shaking. E. coli were treated similarly, but were incubated on ice for the duration of the assay. Each sample was washed, paraformaldehyde-fixed, and re-suspended in PBS. Vitronectin deposition was detected with anti-vitronectin and rabbit anti-goat-Alexafluor488 (Molecular Probes). 5μg/mL DAPI was added to distinguish bacteria from background events. The samples were analyzed as described above.

Serum Sensitivity Assay

E. coli BL21(DE3) harboring pET22b, pJP01, or pJP01-L# derivatives were grown as described above. Bacteria were washed in PBS and approximately 106 colony forming units (cfu) were re-suspended in 200μl of PBS or PBS/50% NHS and incubated for 1 hour at 37°C with agitation. After incubation, samples were serially diluted in PBS and plated on LB-agar plates to determine recovered cfu. Data are presented as the number of bacteria recovered in PBS and PBS/50%NHS after the one-hour incubation period and plotted on a logarithmic scale. Independent triplicate samples were processed for each experimental condition and the experiment was repeated a minimum of three times.

Supplementary Material

Supp Fig S1-S3

Supplemental Figure 1. R. rickettsii Adr1 mediates vitronectin acquisition and serum resistance. A. The R. rickettsii Adr1 homolog of Adr1 is nearly identical to that of R. conorii, which had been characterized throughout this manuscript. Incubation of E. coli expressing a R. rickettsii Adr1 derivative containing an E. coli optimized signal sequence and His6-tag was incubated with NHS, followed by SDS-PAGE separation, and western blotting. IPTG-induced E. coli BL21(pRR7045) and not empty vector or uninduced controls mediated vitronectin (Vn) acquisition as demonstrated by the presence of anti-Vn bands similar to the normal Human serum (NHS) control lane. R. rickettsii Adr1 expression was demonstrated through both anti-Adr1 and anti-His blots. Presence of E. coli in each lane was confirmed by anti-E. coli RNA polymerase α-subunit (RNAP). B. Only E. coli expressing R. rickettsii Adr1 (BL21(pRR7045)+IPTG) are capable of resisting the deleterious effects of NHS. Those bacteria harboring the empty vector (pET22b) or uninduced Adr1 plasmid ((pRR7045)−IPTG) are sensitive to NHS. Statistical comparisons were performed using the student’s t-test.

Supplemental Figure 2. R. typhi Adr1 mediates vitronectin acquisition and serum resistance. A. R. typhi Adr1 (RT815) is 69.3% identical and 80.5% similar to R. conorii Adr1. Incubation of E. coli expressing a R. typhi Adr1 derivative containing an E. coli optimized signal sequence and His6-tag was incubated with NHS, followed by SDS-PAGE separation, and western blotting. IPTG-induced E. coli BL21(pRT815) and not empty vector or uninduced controls mediated vitronectin (Vn) acquisition as demonstrated by the presence of anti-Vn bands similar to the normal Human serum (NHS) control lane. R. typhi Adr1 expression was demonstrated through anti-Adr1 blot. Presence of E. coli in each lane was confirmed by anti-E. coli RNA polymerase α-subunit (RNAP). B. Only E. coli expressing R. typhi Adr1 (BL21(pRT815)+IPTG) are capable of resisting the deleterious effects of NHS. Those bacteria harboring the empty vector (pET22b) or uninduced Adr1 plasmid ((pRT815)−IPTG) are sensitive to NHS. Statistical comparisons were performed using the student’s t-test.

Supplemental Figure 3. R. prowazekii Adr1 mediates vitronectin acquisition and serum resistance. A. R. prowazekii Adr1 (RP827) is 69.5% identical and 79.9% similar to R. conorii Adr1. Incubation of E. coli expressing a R. prowazekii Adr1 derivative containing an E. coli optimized signal sequence and His6-tag was incubated with NHS, followed by SDS-PAGE separation, and western blotting. IPTG-induced E. coli BL21(pRP827) not empty vector (pET22b) or uninduced controls ((pRP827)−IPTG) mediated vitronectin (Vn) acquisition as demonstrated by the presence of anti-Vn bands similar to the normal Human serum (NHS) control lane. R. prowazekii Adr1 expression was demonstrated through anti-His blot. Presence of E. coli in each lane was confirmed by anti-E. coli RNA polymerase α-subunit (RNAP). B. Only E. coli expressing R. prowazekii Adr1 (BL21(pRP827)+IPTG) are capable of resisting the deleterious effects of NHS. Those bacteria harboring the empty vector (pET22b) or uninduced Adr1 plasmid ((pRP827)−IPTG) are sensitive to NHS. Statistical comparisons were performed using the student’s t-test.

Supplemental Table 1. Primers used in this study.

Acknowledgments

The authors would like to thank Marissa Cardwell, Yvonne Chan, Robert Hillman, and William Burke for helpful comments. This work was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Numbers AI072606 and AI103912 (JJM). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References

  1. Ammerman NC, Beier-Sexton M, Azad AF. Laboratory maintenance of Rickettsia rickettsii. Current Protocols in Microbiology. 2008;Chapter 3(Unit 3A):5. doi: 10.1002/9780471729259.mc03a05s11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Anacker RL, McDonald GA, List RH, Mann RE. Neutralizing activity of monoclonal antibodies to heat-sensitive and heat-resistant epitopes of Rickettsia rickettsii surface proteins. Infection and immunity. 1987;55:825–827. doi: 10.1128/iai.55.3.825-827.1987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bartra SS, Styer KL, O’Bryant DM, Nilles ML, Hinnebusch BJ, Aballay A, Plano GV. Resistance of Yersinia pestis to complement-dependent killing is mediated by the Ail outer membrane protein. Infection and immunity. 2008;76:612–622. doi: 10.1128/IAI.01125-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bechah Y, Capo C, Mege JL, Raoult D. Epidemic typhus. Lancet Infectious Diseases. 1998;8:417–426. doi: 10.1016/S1473-3099(08)70150-6. [DOI] [PubMed] [Google Scholar]
  5. Blom AM, Hallstrom T, Riesbeck K. Complement evasion strategies of pathogens-acquisition of inhibitors and beyond. Molecular immunology. 2009;46:2808–2817. doi: 10.1016/j.molimm.2009.04.025. [DOI] [PubMed] [Google Scholar]
  6. Chan YG, Riley SP, Martinez JJ. Adherence to and invasion of host cells by spotted Fever group rickettsia species. Front Microbiol. 2010;1:139. doi: 10.3389/fmicb.2010.00139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chan YGY, Cardwell MM, Hermanas TM, Uchiyama T, Martinez JJ. Rickettial outer-membrane protein B (rOmpB) mediates bacterial invasion through Ku70 in an actin, c-Cbl, clathrin and caveolin 2-dependent manner. Cellular microbiology. 2009;11:629–644. doi: 10.1111/j.1462-5822.2008.01279.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chan YGY, Riley SP, Chen E, Martinez JJ. The molecular basis of immunity to rickettsial infection conferred through outer membrane protein B. Infection and immunity. 2011;79:2303–2313. doi: 10.1128/IAI.01324-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chapman AS, Murphy SM, Demma LJ, Holman RC, Curns AT, McQuiston JH, et al. Rocky mountain spotted fever in the United States, 1997–2002. Annals of the New York Academy of Sciences. 2006;1078:154–155. doi: 10.1196/annals.1374.026. [DOI] [PubMed] [Google Scholar]
  10. Dalton MJ, Clarke MJ, Holman RC, Krebs JW, Fishbein DB, Olson JG, Childs JE. National surveillance for Rocky Mountain spotted fever, 1981–1992: epidemiologic summary and evaluation of risk factors for fatal outcome. The American journal of tropical medicine and hygiene. 1995;52:405–413. doi: 10.4269/ajtmh.1995.52.405. [DOI] [PubMed] [Google Scholar]
  11. Ember JA, Hugli TE. Complement factors and their receptors. Immunopharmacology. 1997;38:3–15. doi: 10.1016/s0162-3109(97)00088-x. [DOI] [PubMed] [Google Scholar]
  12. Fournier PE, El Karkouri K, Leroy Q, Robert C, Giumelli B, Renesto P, et al. Analysis of the Rickettsia africae genome reveals that virulence acquisition in Rickettsia species may be explained by genome reduction. BMC genomics. 2009;10:166. doi: 10.1186/1471-2164-10-166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Helmick CG, Bernard KW, D’Angelo LJ. Rocky Mountain spotted fever: clinical, laboratory, and epidemiological features of 262 cases. The Journal of infectious diseases. 1984;150:480–488. doi: 10.1093/infdis/150.4.480. [DOI] [PubMed] [Google Scholar]
  14. Ho DK, Jarva H, Meri S. Human complement factor H binds to outer membrane protein Rck of Salmonella. Journal of Immunology. 2010;185:1763–1769. doi: 10.4049/jimmunol.1001244. [DOI] [PubMed] [Google Scholar]
  15. Hong H, Patel DR, Tamm LK, van den Berg B. The outer membrane protein OmpW forms an eight-stranded beta-barrel with a hydrophobic channel. The Journal of biological chemistry. 2006;281:7568–7577. doi: 10.1074/jbc.M512365200. [DOI] [PubMed] [Google Scholar]
  16. Jacob-Dubuisson F, Fernandez R, Coutte L. Protein secretion through autotransporter and two-partner pathways. Biochimica et biophysica acta. 2004;1694:235–257. doi: 10.1016/j.bbamcr.2004.03.008. [DOI] [PubMed] [Google Scholar]
  17. Kelley LA, Sternberg MJ. Protein structure prediction on the Web: a case study using the Phyre server. Nature protocols. 2009;4:363–371. doi: 10.1038/nprot.2009.2. [DOI] [PubMed] [Google Scholar]
  18. Lambris JD, Ricklin D, Geisbrecht BV. Complement evasion by human pathogens. Nature reviews Microbiology. 2008;6:132–142. doi: 10.1038/nrmicro1824. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Li R, Ren M, Chen N, Luo M, Zhang Z, Wu J. Vitronectin increases vascular permeability by promoting VE-cadherin internalization at cell junctions. PloS one. 2012;7:e37195. doi: 10.1371/journal.pone.0037195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Meri S, Jarva H. Complement regulation. Vox sanguinis. 1998;74(Suppl 2):291–302. doi: 10.1111/j.1423-0410.1998.tb05434.x. [DOI] [PubMed] [Google Scholar]
  21. Nikaido H. Isolation of outer membranes. Methods in enzymology. 1994;235:225–234. doi: 10.1016/0076-6879(94)35143-0. [DOI] [PubMed] [Google Scholar]
  22. Pangburn MK, Schreiber RD, Muller-Eberhard HJ. Human complement C3b inactivator: isolation, characterization, and demonstration of an absolute requirement for the serum protein beta1H for cleavage of C3b and C4b in solution. The Journal of experimental medicine. 1977;146:257–270. doi: 10.1084/jem.146.1.257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Pautsch A, Schulz GE. Structure of the outer membrane protein A transmembrane domain. Nature structural biology. 1998;5:1013–1017. doi: 10.1038/2983. [DOI] [PubMed] [Google Scholar]
  24. Peterson CB. Binding sites on native and multimeric vitronectin exhibit similar affinity for heparin the influence of self-association and multivalence on ligand binding. Trends in cardiovascular medicine. 1998;8:124–131. doi: 10.1016/S1050-1738(97)00136-9. [DOI] [PubMed] [Google Scholar]
  25. Preissner KT, Reuning U. Vitronectin in vascular context: facets of a multitalented matricellular protein. Seminars in thrombosis and hemostasis. 2011;37:408–424. doi: 10.1055/s-0031-1276590. [DOI] [PubMed] [Google Scholar]
  26. Raoult D, Roux V. Rickettsioses as paradigms of new or emerging infectious diseases. Clin Microbiol Rev. 1997;10:694–719. doi: 10.1128/cmr.10.4.694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Renesto P, Samson L, Ogata H, Azza S, Fourquet P, Gorvel JP, et al. Identification of two putative rickettsial adhesins by proteomic analysis. Research in Microbiology. 2006;157:605–612. doi: 10.1016/j.resmic.2006.02.002. [DOI] [PubMed] [Google Scholar]
  28. Riley SP, Patterson JL, Martinez JJ. The rickettsial OmpB beta-peptide of Rickettsia conorii is sufficient to facilitate factor H-mediated serum resistance. Infection and immunity. 2012;80:2735–2743. doi: 10.1128/IAI.00349-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Singh B, Su YC, Riesbeck K. Vitronectin in bacterial pathogenesis: a host protein used in complement escape and cellular invasion. Molecular Microbiology. 2010;78:545–560. doi: 10.1111/j.1365-2958.2010.07373.x. [DOI] [PubMed] [Google Scholar]
  30. Tegla CA, Cudrici C, Patel S, Trippe R, 3rd, Rus V, Niculescu F, Rus H. Membrane attack by complement: the assembly and biology of terminal complement complexes. Immunologic research. 2011;51:45–60. doi: 10.1007/s12026-011-8239-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Thanassi DG, Saulino ET, Lombardo MJ, Roth R, Heuser J, Hultgren SJ. The PapC usher forms an oligomeric channel: implications for pilus biogenesis across the outer membrane. Proceedings of the National Academy of Sciences of the United States of America. 1998;95:3146–3151. doi: 10.1073/pnas.95.6.3146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Tsuruta Y, Park YJ, Siegal GP, Liu G, Abraham E. Involvement of vitronectin in lipopolysaccaride-induced acute lung injury. J Immunol. 2007;179:7079–7086. doi: 10.4049/jimmunol.179.10.7079. [DOI] [PubMed] [Google Scholar]
  33. Vandeputte-Rutten L, Bos MP, Tommassen J, Gros P. Crystal structure of Neisserial surface protein A (NspA), a conserved outer membrane protein with vaccine potential. The Journal of biological chemistry. 2003;278:24825–24830. doi: 10.1074/jbc.M302803200. [DOI] [PubMed] [Google Scholar]
  34. Vogt J, Schulz GE. The structure of the outer membrane protein OmpX from Escherichia coli reveals possible mechanisms of virulence. Structure. 1999;7:1301–1309. doi: 10.1016/s0969-2126(00)80063-5. [DOI] [PubMed] [Google Scholar]
  35. Walker DH, Herrero-Herrero JI, Ruiz-Beltran R, Bullon-Sopelana A, Ramos-Hidalgo A. The pathology of fatal Mediterranean spotted fever. American Journal of Clinical Pathology. 1987;87:669–672. doi: 10.1093/ajcp/87.5.669. [DOI] [PubMed] [Google Scholar]
  36. Walker TS, Winkler HH. Penetration of cultured mouse fibroblasts (L cells) by Rickettsia prowazeki. Infect Immun. 1978;22:200–208. doi: 10.1128/iai.22.1.200-208.1978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Walport MJ. Complement. First of two parts. The New England journal of medicine. 2001a;344:1058–1066. doi: 10.1056/NEJM200104053441406. [DOI] [PubMed] [Google Scholar]
  38. Walport MJ. Complement. Second of two parts. The New England journal of medicine. 2001b;344:1140–1144. doi: 10.1056/NEJM200104123441506. [DOI] [PubMed] [Google Scholar]
  39. Weiler JM, Daha MR, Austen KF, Fearon DT. Control of the amplification convertase of complement by the plasma protein beta1H. Proceedings of the National Academy of Sciences of the United States of America. 1976;73:3268–3272. doi: 10.1073/pnas.73.9.3268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zipfel PF, Smith RJ, Skerka C. Factor I and factor H deficiency in renal diseases: similar defects in the fluid phase have a different outcome at the surface of the glomerular basement membrane. Nephrology, dialysis, transplantation: official publication of the European Dialysis and Transplant Association - European Renal Association. 2009;24:385–387. doi: 10.1093/ndt/gfn652. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Fig S1-S3

Supplemental Figure 1. R. rickettsii Adr1 mediates vitronectin acquisition and serum resistance. A. The R. rickettsii Adr1 homolog of Adr1 is nearly identical to that of R. conorii, which had been characterized throughout this manuscript. Incubation of E. coli expressing a R. rickettsii Adr1 derivative containing an E. coli optimized signal sequence and His6-tag was incubated with NHS, followed by SDS-PAGE separation, and western blotting. IPTG-induced E. coli BL21(pRR7045) and not empty vector or uninduced controls mediated vitronectin (Vn) acquisition as demonstrated by the presence of anti-Vn bands similar to the normal Human serum (NHS) control lane. R. rickettsii Adr1 expression was demonstrated through both anti-Adr1 and anti-His blots. Presence of E. coli in each lane was confirmed by anti-E. coli RNA polymerase α-subunit (RNAP). B. Only E. coli expressing R. rickettsii Adr1 (BL21(pRR7045)+IPTG) are capable of resisting the deleterious effects of NHS. Those bacteria harboring the empty vector (pET22b) or uninduced Adr1 plasmid ((pRR7045)−IPTG) are sensitive to NHS. Statistical comparisons were performed using the student’s t-test.

Supplemental Figure 2. R. typhi Adr1 mediates vitronectin acquisition and serum resistance. A. R. typhi Adr1 (RT815) is 69.3% identical and 80.5% similar to R. conorii Adr1. Incubation of E. coli expressing a R. typhi Adr1 derivative containing an E. coli optimized signal sequence and His6-tag was incubated with NHS, followed by SDS-PAGE separation, and western blotting. IPTG-induced E. coli BL21(pRT815) and not empty vector or uninduced controls mediated vitronectin (Vn) acquisition as demonstrated by the presence of anti-Vn bands similar to the normal Human serum (NHS) control lane. R. typhi Adr1 expression was demonstrated through anti-Adr1 blot. Presence of E. coli in each lane was confirmed by anti-E. coli RNA polymerase α-subunit (RNAP). B. Only E. coli expressing R. typhi Adr1 (BL21(pRT815)+IPTG) are capable of resisting the deleterious effects of NHS. Those bacteria harboring the empty vector (pET22b) or uninduced Adr1 plasmid ((pRT815)−IPTG) are sensitive to NHS. Statistical comparisons were performed using the student’s t-test.

Supplemental Figure 3. R. prowazekii Adr1 mediates vitronectin acquisition and serum resistance. A. R. prowazekii Adr1 (RP827) is 69.5% identical and 79.9% similar to R. conorii Adr1. Incubation of E. coli expressing a R. prowazekii Adr1 derivative containing an E. coli optimized signal sequence and His6-tag was incubated with NHS, followed by SDS-PAGE separation, and western blotting. IPTG-induced E. coli BL21(pRP827) not empty vector (pET22b) or uninduced controls ((pRP827)−IPTG) mediated vitronectin (Vn) acquisition as demonstrated by the presence of anti-Vn bands similar to the normal Human serum (NHS) control lane. R. prowazekii Adr1 expression was demonstrated through anti-His blot. Presence of E. coli in each lane was confirmed by anti-E. coli RNA polymerase α-subunit (RNAP). B. Only E. coli expressing R. prowazekii Adr1 (BL21(pRP827)+IPTG) are capable of resisting the deleterious effects of NHS. Those bacteria harboring the empty vector (pET22b) or uninduced Adr1 plasmid ((pRP827)−IPTG) are sensitive to NHS. Statistical comparisons were performed using the student’s t-test.

Supplemental Table 1. Primers used in this study.

NIHMS545750-supplement-Supp_Fig_S1-S3.pdf (648.9KB, pdf)

RESOURCES