Heterologous Expression of Bartonella Adhesin A in Escherichia coli by Exchange of Trimeric Autotransporter Adhesin Domains Results in Enhanced Adhesion Properties and a Pathogenic Phenotype

Human-pathogenic Bartonella henselae causes cat scratch disease and vasculoproliferative disorders. An important pathogenicity factor of B. henselae is the trimeric autotransporter adhesin (TAA) Bartonella adhesin A (BadA), which is modularly constructed, consisting of a head, a long and repetitive neck-stalk module, and a membrane anchor. BadA is involved in bacterial autoagglutination, binding to extracellular matrix proteins and host cells, and in proangiogenic reprogramming. The slow growth of B. henselae and limited tools for genetic manipulation are obstacles for detailed examination of BadA and its domains. Here, we established a recombinant expression system for BadA mutants in Escherichia coli allowing functional analysis of particular BadA domains. Using a BadA mutant lacking 21 neck-stalk repeats (BadA HN23), the BadA HN23 signal sequence was exchanged with that of E. coli OmpA, and the BadA membrane anchor was additionally replaced with that of Yersinia adhesin A (YadA). Constructs were cloned in E. coli, and hybrid protein expression was detected by immunoblotting, fluorescence microscopy, and flow cytometry. Functional analysis revealed that BadA hybrid proteins mediate autoagglutination and binding to collagen and endothelial cells. In vivo, expression of this BadA construct correlated with higher pathogenicity of E. coli in a Galleria mellonella infection model.     

Structure of the head of the Bartonella adhesin BadA

Trimeric autotransporter adhesins (TAAs) are a major class of proteins by which pathogenic proteobacteria adhere to their hosts. Prominent examples include Yersinia YadA, Haemophilus Hia and Hsf, Moraxella UspA1 and A2, and Neisseria NadA. TAAs also occur in symbiotic and environmental species and presumably represent a general solution to the problem of adhesion in proteobacteria. The general structure of TAAs follows a head-stalk-anchor architecture, where the heads are the primary mediators of attachment and autoagglutination. In the major adhesin of Bartonella henselae, BadA, the head consists of three domains, the N-terminal of which shows strong sequence similarity to the head of Yersinia YadA. The two other domains were not recognizably similar to any protein of known structure. We therefore determined their crystal structure to a resolution of 1.1 A. Both domains are beta-prisms, the N-terminal one formed by interleaved, five-stranded beta-meanders parallel to the trimer axis and the C-terminal one by five-stranded beta-meanders orthogonal to the axis. Despite the absence of statistically significant sequence similarity, the two domains are structurally similar to domains from Haemophilus Hia, albeit in permuted order. Thus, the BadA head appears to be a chimera of domains seen in two other TAAs, YadA and Hia, highlighting the combinatorial evolutionary strategy taken by pathogens.     

Bartonella henselae trimeric autotransporter adhesin BadA expression interferes with effector translocation by the VirB/D4 type IV secretion system

The Gram‐negative, zoonotic pathogen Bartonella henselae is the aetiological agent of cat scratch disease, bacillary angiomatosis and peliosis hepatis in humans. Two pathogenicity factors of B. henselae – each displaying multiple functions in host cell interaction – have been characterized in greater detail: the trimeric autotransporter Bartonella adhesin A (BadA) and the type IV secretion system VirB/D4 (VirB/D4 T4SS). BadA mediates, e.g. binding to fibronectin (Fn), adherence to endothelial cells (ECs) and secretion of vascular endothelial growth factor (VEGF). VirB/D4 translocates several Bartonella effector proteins (Beps) into the cytoplasm of infected ECs, resulting, e.g. in uptake of bacterial aggregates via the invasome structure, inhibition of apoptosis and activation of a proangiogenic phenotype. Despite this knowledge of the individual activities of BadA or VirB/D4 it is unknown whether these major virulence factors affect each other in their specific activities. In this study, expression and function of BadA and VirB/D4 were analysed in a variety of clinical B. henselae isolates. Data revealed that mostisolates have lost expression of either BadA or VirB/D4 during in vitro passages. However, the phenotypic effects of coexpression of both virulence factors was studied in one clinical isolate that was found to stably coexpress BadA and VirB/D4, as well as by ectopic expression of BadA in a strain expressing VirB/D4 but not BadA. BadA, which forms a dense layer on the bacterial surface, negatively affected VirB/D4‐dependent Bep translocation and invasome formation by likely preventing close contact between the bacterial cell envelope and the host cell membrane. In contrast, BadA‐dependent Fn binding, adhesion to ECs and VEGF secretion were not affected by a functional VirB/D4 T4SS. The obtained data imply that the essential virulence factors BadA and VirB/D4 are likely differentially expressed during different stages of the infection cycle of Bartonella.     

Analysis of the BadA stalk from Bartonella henselae reveals domain-specific and domain-overlapping functions in the host cell infection process

Human pathogenic Bartonella henselae cause cat scratch disease and vasculoproliferative disorders. An important pathogenicity factor of B. henselae is the trimeric autotransporter adhesin Bartonella adhesin A (BadA) which is modularly constructed and consists of a head, a long and repetitive neck‐stalk module with 22 repetitive neck/stalk repeats and a membrane anchor. The BadA head is crucial for bacterial adherence to host cells, binding to several extracellular matrix proteins and for the induction of vascular endothelial growth factor (VEGF) secretion. Here, we analysed the biological role of the BadA stalk in the infection process in greater detail. For this purpose, BadA head‐bearing and headless deletion mutants with different lengths (containing one or four neck/stalk repeats in the neck‐stalk module) were produced and functionally analysed for their ability to bind to fibronectin, collagen and endothelial cells and to induce VEGF secretion. Whereas a head‐bearing short version (one neck/stalk element) of BadA lacks exclusively fibronectin binding, a substantially truncated headless BadA mutant was deficient for all of these biological functions. The expression of a longer headless BadA mutant (four neck/stalk repeats) restored fibronectin and collagen binding, adherence to host cells and the induction of VEGF secretion. Our data suggest that (i) the stalk of BadA is exclusively responsible for fibronectin binding and that (ii) both the head and stalk of BadA mediate adherence to collagen and host cells and the induction of VEGF secretion. This indicates overlapping functions of the BadA head and stalk.     

Staining of Bartonella henselae with carboxyfluorescein diacetate succinimidyl ester for tracking infection in erythrocytes and epithelial cells

Bartonella infection (Bartonella henselae in particular) is responsible for a widening spectrum of human diseases. The persistent colonization of erythrocytes is a feature of Bartonella infection. Endothelial and epithelial cells are also widely used to study the pathogenesis of bartonellosis in vitro. Exploring a convenient method for visualizing the bacillus without affecting infectivity would be very interesting. Carboxyfluorescein diacetate succinimidyl ester (CFSE) has been previously used for staining several bacterial species to study their adhesion to host cells. The present study demonstrated the efficiency and safety of using CFSE in staining B. henselae. The staining of bacillus-invaded erythrocytes and epithelial cells in vitro successfully allowed for flow cytometry and confocol microscopy analyses. Parallel tests using untreated bacteria confirmed that CFSE staining did not result in side effects on the infectivity of B. henselae. Labeling Bartonella with CFSE is a valuable method for studying the bacteria-host interaction.     

Interaction of Bartonella henselae with endothelial cells results in rapid bacterial rRNA synthesis and replication

Bartonella henselae is a slow-growing microorganism and the causative pathogen of bacillary angiomatosis in man. Here, we analysed how interaction of B. henselae with endothelial cells might affect bacterial growth. For this purpose, bacterial rRNA production and ribosome content was determined by fluorescence in situ hybridization (FISH) using rRNA-targeted fluorescence-labelled oligonucleotide probes. B. henselae grown on agar plates showed no detectable rRNA content by means of FISH, whereas B. henselae co-cultured with endothelial cells showed a rapid increase of rRNA production within the first 18 h after inoculation. The increased rRNA synthesis was paralleled by a ∼1000-fold intracellular bacterial replication, whereas bacteria grown on agar base showed only a ∼10-fold replication within the first 48 h of culture. Pretreatment of host cells with paraformaldehyde prevented adhesion, invasion, intracellular replication and bacterial rRNA synthesis of B. henselae . In contrast, inhibition of host cell protein synthesis by cycloheximide did not affect bacterial adhesion and invasion, but prevented intracellular replication although bacterial rRNA content was increased. Inhibition of actin polymerization by cytochalasin D did not affect adhesion, invasion, increased rRNA content or intracellular replication of B. henselae. These results demonstrate that rRNA synthesis and replication of B. henselae is promoted by viable host cells with intact de novo protein synthesis.     

Evidence of a leading role for VEGF in Bartonella henselae-induced endothelial cell proliferations

Bartonella henselae causes the vasculoproliferative disorders bacillary angiomatosis (BA) and bacillary peliosis (BP). The pathomechanisms of these tumorous proliferations are unknown. Our results suggest a novel bacterial two-step pathogenicity strategy, in which the pathogen triggers growth factor production for subsequent proliferation of its own host cells. In fact, B. henselae induces host cell production of the angiogenic factor vascular endothelial growth factor (VEGF), leading to proliferation of endothelial cells. The presence of B. henselae pili was associated with host cell VEGF production, as a Pil- mutant of B. henselae was unable to induce VEGF production. In turn, VEGF-stimulated endothelial cells promoted the growth of B. henselae. Immunohistochemistry for VEGF in specimens from patients with BA or BP revealed increased VEGF expression in vivo. These findings suggest a novel bacteria-dependent mechanism of tumour growth.     

Differential effects of Bartonella henselae on human and feline macro- and micro-vascular endothelial cells

Bartonella henselae, a zoonotic agent, induces tumors of endothelial cells (ECs), namely bacillary angiomatosis and peliosis in immunosuppressed humans but not in cats. In vitro studies on ECs represent to date the only way to explore the interactions between Bartonella henselae and vascular endothelium. However, no comparative study of the interactions between Bartonella henselae and human (incidental host) ECs vs feline (reservoir host) ECs has been carried out because of the absence of any available feline endothelial cell lines.To this purpose, we have developed nine feline EC lines which allowed comparing the effects of Bartonella strains on human and feline micro-vascular ECs representative of the infection development sites such as skin, versus macro-vascular ECs, such as umbilical vein.Our model revealed intrinsic differences between human (Human Skin Microvascular ECs -HSkMEC and Human Umbilical Vein ECs - iHUVEC) and feline ECs susceptibility to Bartonella henselae infection.While no effect was observed on the feline ECs upon Bartonella henselae infection, the human ones displayed accelerated angiogenesis and wound healing.Noticeable differences were demonstrated between human micro- and macro-vasculature derived ECs both in terms of pseudo-tube formation and healing. Interestingly, Bartonella henselae effects on human ECs were also elicited by soluble factors.Neither Bartonella henselae-infected Human Skin Microvascular ECs clinically involved in bacillary angiomatosis, nor feline ECs increased cAMP production, as opposed to HUVEC.Bartonella henselae could stimulate the activation of Vascular Endothelial Growth Factor Receptor-2 (VEGFR-2) in homologous cellular systems and trigger VEGF production by HSkMECs only, but not iHUVEC or any feline ECs tested.These results may explain the decreased pathogenic potential of Bartonella henselae infection for cats as compared to humans and strongly suggest that an autocrine secretion of VEGF by human skin endothelial cells might induce their growth and ultimately lead to bacillary angiomatosis formation.     

Unusual trafficking pattern of Bartonella henselae -containing vacuoles in macrophages and endothelial cells

Bartonella henselae, the agent of cat-scratch disease and vasculoproliferative disorders in humans, is a fastidious facultative intracellular pathogen, whose interaction with macrophages and endothelial cells (ECs) is crucial in the pathogenesis of these diseases. However, little is known about the subcellular compartment in which B. henselae resides. Two hours after infection of murine macrophages and human ECs, the majority of B. henselae-containing vacuoles (BCVs) lack typical endocytic marker proteins, fail to acidify, and do not fuse with lysosomes, suggesting that B. henselae resides in a non-endocytic compartment. In contrast to human umbilical vein endothelial cells, bacterial death and lysosomal fusion with BCVs is apparent in J774A.1 macrophages at 24 h. This phenomenon of delayed lysosomal fusion requires bacterial viability, and is confined to the BCV itself. Using magnetic selection, we enriched for transposon-mutagenized B. henselae trapped in lysosomes of macrophages 2 h after infection. Genes affected appear to be relevant to the intracellular lifestyle in macrophages and ECs and include some previously implicated in Bartonella pathogenicity. We conclude that B. henselae has a specific capacity to actively avoid the host endocytic pathway after entry of macrophages and ECs, from within a specialized non-endocytic membrane-bound vacuole.     

Prevalence of Bartonella henselae in stray and domestic cats in different Italian areas: evaluation of the potential risk of transmission of Bartonella to humans

Bartonella henselae is the major etiological agent of Cat Scratch Disease in humans. Cats act as the natural reservoir of B. henselae and can transmit the infection to humans by bite or scratch. The diffusion of B. henselae was evaluated by seroprevalence and bacteremic status in different stray cat populations located in nine areas of Northern Italy. A total of 1585 cats were tested by blood culture and 361 (23%) resulted bacteremic; 1416 out off 1585 cats were also tested for Bartonella henselae antibodies and 553 (39%) resulted seropositive. The molecular typing of the isolates showed that 26% of bacteremic cats were infected with B. henselae type I, 52% with B. henselae type II, 16% were co-infected with both and 5% infected with B. Clarridgeiae. Moreover 165 domestic cats were tested by blood culture and serological test (IFA test cut-off: 1:64). 35 cats (21%) resulted bacteremic and 49 (43.5%) were seropositive. The molecular typing of the Bartonella isolates of the domestic cats showed that 45% of bacteremic cats were infected with B. henselae type I, 36.5% with B. henselae type II, 12% were coinfected with both and 6% infected with B. Clarridgeiae. For a completely evaluation of health status of the cat for B. henselae infection, the authors suggest both blood culture and serological tests. Nevertheless a nonbacteremic cat with positive serology result should be reevaluated for possible recurrent bacteremia.     

The VirB type IV secretion system of Bartonella henselae mediates invasion, proinflammatory activation and antiapoptotic protection of endothelial cells

Bartonella henselae is an arthropod-borne zoonotic pathogen causing intraerythrocytic bacteraemia in the feline reservoir host and a broad range of clinical manifestations in incidentally infected humans. Remarkably, B. henselae can specifically colonize the human vascular endothelium, resulting in inflammation and the formation of vasoproliferative lesions known as bacillary angiomatosis and bacillary peliosis. Cultured human endothelial cells provide an in vitro system to study this intimate interaction of B. henselae with the vascular endothelium. However, little is known about the bacterial virulence factors required for this pathogenic process. Recently, we identified the type IV secretion system (T4SS) VirB as an essential pathogenicity factor in Bartonella, required to establish intraerythrocytic infection in the mammalian reservoir. Here, we demonstrate that the VirB T4SS also mediates most of the virulence attributes associated with the interaction of B. henselae during the interaction with human endothelial cells. These include: (i) massive rearrangements of the actin cytoskeleton, resulting in the formation of bacterial aggregates and their internalization by the invasome structure; (ii) nuclear factor kappaB-dependent proinflammatory activation, leading to cell adhesion molecule expression and chemokine secretion, and (iii) inhibition of apoptotic cell death, resulting in enhanced endothelial cell survival. Moreover, we show that the VirB system mediates cytostatic and cytotoxic effects at high bacterial titres, which interfere with a potent VirB-independent mitogenic activity. We conclude that the VirB T4SS is a major virulence determinant of B. henselae, required for targeting multiple endothelial cell functions exploited by this vasculotropic pathogen.     

Isolation of bacteriophages from Bartonella vinsonii subsp. berkhoffii and the characterization of Pap31 gene sequences from bacterial and phage DNA

Bacteriophages enhance bacterial survival, facilitate bacterial adaptation to new environmental conditions, assist in the adaptation to a new host species, and enhance bacterial evasion or inactivation of host defense mechanisms. We describe the detection and purification of a novel tailed bacteriophage from Bartonella vinsonii subsp. berkhoffii, which was previously described as a bacteriophage-negative species. We also compare B. vinsonii subsp. berkhoffi Pap31 bacteriophage gene sequences to B. henselae (Houston I), and B. quintana (Fuller) bacteriophage Pap31 sequences. Negative staining electron microscopy of log phase culturesof B. vinsonii subsp. berkhoffii identified bacteriophages, possessing a 50-nm icosahedric head diameter and a 60- to 80-nm contractile tail. Sequence analysis of the bacteriophage Pap31 gene from B. vinsonii subsp. berkhoffii showed three consensus sequences and a 12-bp insertion when compared with Pap31 gene sequences from B. henselae (Houston I) and B. quintana (Fuller) bacteriophages. Isolation of B. vinsonii subsp. berkhoffii bacteriophages containing a Pap31 gene suggests that this heme-binding protein gene might play an important role in bacterial virulence through the genetic exchange of DNA within this subspecies. Defining phage-associated genes may also contribute to the enhanced understanding of the evolutionary relationships among members of the genus Bartonella.     

Genome rearrangements, deletions, and amplifications in the natural population of Bartonella henselae

Cats are the natural host for Bartonella henselae, an opportunistic human pathogen and the agent of cat scratch disease. Here, we have analyzed the natural variation in gene content and genome structure of 38 Bartonella henselae strains isolated from cats and humans by comparative genome hybridizations to microarrays and probe hybridizations to pulsed-field gel electrophoresis (PFGE) blots. The variation in gene content was modest and confined to the prophage and the genomic islands, whereas the PFGE analyses indicated extensive rearrangements across the terminus of replication with breakpoints in areas of the genomic islands. We observed no difference in gene content or structure between feline and human strains. Rather, the results suggest multiple sources of human infection from feline B. henselae strains of diverse genotypes. Additionally, the microarray hybridizations revealed DNA amplification in some strains in the so-called chromosome II-like region. The amplified segments were centered at a position corresponding to a putative phage replication initiation site and increased in size with the duration of cultivation. We hypothesize that the variable gene pool in the B. henselae population plays an important role in the establishment of long-term persistent infection in the natural host by promoting antigenic variation and escape from the host immune response.     

Characterization of the genome composition of Bartonella koehlerae by microarray comparative genomic hybridization profiling

Bartonella henselae is present in a wide range of wild and domestic feline hosts and causes cat-scratch disease and bacillary angiomatosis in humans. We have estimated here the gene content of Bartonella koehlerae, a novel species isolated from cats that was recently identified as an agent of human endocarditis. The investigation was accomplished by comparative genomic hybridization (CGH) to a microarray constructed from the sequenced 1.93-Mb genome of B. henselae. Control hybridizations of labeled DNA from the human pathogen Bartonella quintana with a reduced genome of 1.58 Mb were performed to evaluate the accuracy of the array for genes with known levels of sequence divergence. Genome size estimates of B. koehlerae by pulsed-field gel electrophoresis matched that calculated by the CGH, indicating a genome of 1.7 to 1.8 Mb with few unique genes. As in B. quintana, sequences in the prophage and the genomic islands were reported absent in B. koehlerae. In addition, sequence variability was recorded in the chromosome II-like region, where B. koehlerae showed an intermediate retention pattern of both coding and noncoding sequences. Although most of the genes missing in B. koehlerae are also absent from B. quintana, its phylogenetic placement near B. henselae suggests independent deletion events, indicating that host specificity is not solely attributed to genes in the genomic islands. Rather, the results underscore the instability of the genomic islands even within bacterial populations adapted to the same host-vector system, as in the case of B. henselae and B. koehlerae.     

P26-based serodiagnosis for Bartonella spp. infection in cats

Bartonella henselae P26 has been identified as an immunodominant antigen expressed during feline infection. We used antisera from cats experimentally infected with B. henselae (n = 6), B. clarridgeiae (n = 4), or B. koehlerae (n = 2) and from a sample of naturally infected cats (B. henselae, n = 34; B. clarridgeiae, n = 1) to evaluate recombinant P26 (rP26) as a serodiagnostic antigen. Immunoblots using antisera from cats infected with B. henselae and B. clarridgeiae reacted strongly with rP26, whereas B. koehlerae antisera did not. A capture ELISA was designed to evaluate the kinetics of rP26 IgG in sera from experimentally infected cats. For B. henselae and B. clarridgeiae antisera, the kinetic profiles of reactivity were similar for rP26 capture ELISA and Bartonella spp. indirect fluorescence assay. However, for B. koehlerae antisera, reactivity in rP26 capture ELISA was consistently low. The serodiagnostic potential of rP26 capture ELISA was evaluated using sera from cats with known Bartonella sp. exposure histories. All 24 (100%) uninfected cats were seronegative, and 33 of 35 (94.3%) cats bacteremic for Bartonella spp. were seropositive. We propose that rP26-based serology can serve as a useful adjunct tool for the diagnosis of feline infection with B. henselae and B. clarridgeiae, but it may not be useful for feline infection with B. koehlerae.     

A genome-wide study of recombination rate variation in Bartonella henselae

ABSTRACT: BACKGROUND: Rates of recombination vary by three orders of magnitude in bacteria but the reasons for this variation is unclear. We performed a genome-wide study of recombination rate variation among genes in the intracellular bacterium Bartonella henselae, which has among the lowest estimated ratio of recombination relative to mutation in prokaryotes. RESULTS: The 1.9 Mb genomes of B. henselae strains IC11, UGA10 and Houston-1 genomes showed only minor gene content variation. Nucleotide sequence divergence levels were less than 1% and the relative rate of recombination to mutation was estimated to 1.1 for the genome overall. Four to eight segments per genome presented significantly enhanced divergences, the most pronounced of which were the virB and trw gene clusters for type IV secretion systems that play essential roles in the infection process. Consistently, multiple recombination events were identified inside these gene clusters. High recombination frequencies were also observed for a gene putatively involved in iron metabolism. A phylogenetic study of this gene in 80 strains of Bartonella quintana, B. henselae and B. grahamii indicated different population structures for each species and revealed horizontal gene transfers across Bartonella species with different host preferences. CONCLUSIONS: Our analysis has shown little novel gene acquisition in B. henselae, indicative of a closed pan-genome, but higher recombination frequencies within the population than previously estimated. We propose that the dramatically increased fixation rate for recombination events at gene clusters for type IV secretion systems is driven by selection for sequence variability.     

Bartonella entry mechanisms into mammalian host cells

The Gram-negative genus Bartonella comprises arthropod-borne pathogens that typically infect mammals in a host-specific manner. Bartonella bacilliformis and Bartonella quintana are human-specific pathogens, while several zoonotic bartonellae specific for diverse animal hosts infect humans as an incidental host. Clinical manifestations of Bartonella infections range from mild symptoms to life-threatening disease. Following transmission by blood-sucking arthropods or traumatic contact with infected animals, bartonellae display sequential tropisms towards endothelial and possibly other nucleated cells and erythrocytes, the latter in a host-specific manner. Attachment to the extracellular matrix (ECM) and to nucleated cells is mediated by surface-exposed bacterial adhesins, in particular trimeric autotransporter adhesins (TAAs). The subsequent engulfment of the pathogen into a vacuolar structure follows a unique series of events whereby the pathogen avoids the endolysosomal compartments. For Bartonella henselae and assumingly most other species, the infection process is aided at different steps by Bartonella effector proteins (Beps). They are injected into host cells through the type IV secretion system (T4SS) VirB/D4 and subvert host cellular functions to favour pathogen uptake. Bacterial binding to erythrocytes is mediated by Trw, another T4SS, in a strictly host-specific manner, followed by pathogen-forced uptake involving the IalB invasin and subsequent replication and persistence within a membrane-bound intra-erythrocytic compartment.     

Bacteriological and molecular identification of Bartonella species in cats from different regions of China

With the improvements in diagnostic techniques, Bartonella henselae (B. henselae) infection has recently been recognized to cause a widening spectrum of diseases. Cats are the natural reservoir hosts of B. henselae. The current study aims to investigate the prevalence of B. henselae infection in the cat populations in China. Polymerase chain reaction (PCR) and bacterial cultures confirm that 12.7% of the tested cats were positive for the infection. Old age and outdoor exposure were statistically associated with the infection. Multilocus sequence typing and eBURST analysis of the cat isolates collected in the present study show that 65.4% of the isolates belong to sequence type 1 (ST1). Three new STs (ST16-18) were identified in Midwestern China. These results may aid our understanding of the population structure of B. henselae in China and the relationship between human and cat strains in subsequent studies.