Bordetella pertussis is the bacterium that causes pertussis, otherwise known asor whooping cough . Despite vaccination, incidence of pertussis cases have been growing over the last two decades. Current vaccines lack the ability to give long-lasting immunity and must be improved. Complement evasion molecules would make good candidates for vaccine components, because the complement system is essential in the killing of B. pertussis. The complement system is a first line of defense against colonization of bacteria in host tissue. Many bacteria employ complement inhibitors or attract human complement inhibitors to their surface. B. pertussis is no exception on this, however not much is known about the way B. pertussis evades the complement system . Prior to this study a few potential complement inhibitors of B. pertussis were identified. In this study we seek to characterize these proteins in a series of functional immune assays. Under these experimental conditions, we were not able to determine the function of the potential complement evasion molecules.
Bordetella pertussis is a Gram-negative bacterium and the causative agent of pertussis, also called whooping cough. Pertussis is a human-restricted disease of the respiratory tract and highly contagious ( en is dit zo? REF). Regular bouts of coughing produce airborne droplets through which B. pertussis can be transmitted (Jongerius et al. 2014). The disease was on the verge of eradication due to worldwide vaccination, but has been re-emerging in the last two decades. In 2008 there were 16 million cases of pertussis globally as estimated by the World Health Organisation. Possible explanations for the re-emergence are waning of immunity – since in the 1990s a different vaccine was introduced giving less enduring protection – or bacterial strain adaption (Mooi et al. 2014). Like many pathogens B. pertussis developed strategies to evade or repress the host response for successful colonization (REF). Little is known about the mechanisms by which B. pertussis escapes clearance by the complement system, a first line of defense in the immune response. Identification and characterization of these immune evasion molecules may lead to the discovery of new targets for anti-inflammatory drugs or new components for vaccines. Here, we try to characterize several potential complement inhibitors of B. pertussis, identified by phage display and genome analysis.
Bordetella pertussis virulence factors
B. pertussis expresses a variety of virulence factors that enable the bacteria to colonize the upper respiratory tract. The transcription of the majority of these virulence factors are regulated by the activity of the proteins BvgA and BvgS. BvgAS is a two-component system that controls gene expression in response to changing environmental conditions, such as temperature (Fedele et al. 2014). After inhalation the bacteria adhere to the ciliated epithelial cells of the larynx, trachea and bronchi, where they produce secreted toxins and membrane-bound molecules that contribute to the adherence (REF). The toxins damage the mucous layer of the respiratory tract, contributing to the pathogenesis of pertussis. For example tracheal cytotoxin (TCT) is proposed to cause ciliostasis, impairment of ciliary movement, which may explain the intense coughing that characterizes pertussis, as a way to clear excessive mucus (Jongerius et al. 2014). Other secreted toxins include pertussis toxin (PT) and adenylate cyclase toxin (ACT), which are toxic to host cells including neutrophils, monocytes and lymphocytes (Jongerius et al. 2014 ). Successful colonization is both dependent on the capacity to adhere to cells in the respiratory tract and the ability to ward off the immune response. PT and ACT both repress the immune response by targeting airway resident macrophages and neutrophil recruitment to the airways (Carbonetti et al. 2010). The membrane-bound adhesins include fimbriae (Fim), filamentous hemagglutinin (FHA) and pertactin (PRN), which in addition to facilitating the first step of infection are also suggested to suppress the initial inflammatory response to the infection (Melvin et al. 2014). For example, FHA-deficient Bordetella strains were shown to induce an increased production of pro-inflammatory cytokines, such as IL-17, and increased recruitment of neutrophils to the site of infection when compared to wild-type bacteria (Henderson et al. 2012). In addition to the immunomodulating properties of B. pertussis virulence factors, direct binding to components of the immune system, such as the complement system, is accomplished in order to reduce or inhibit the host immune response. This essential step in bacterial pathogenesis is called immune evasion (Rooijakkers et al. 2005). B. pertussis is known to express a small variety of proteins that affect complement-mediated killing: tracheal colonization factor (Tcf), Bordetella resistance to killing A (BrkA) and the autotransporter Vag8 (Jongerius et al. 2014).
The complement system
The complement system is part of innate immunity and an important host defense mechanism against invading pathogens. The complement system consists of approximately 40 proteins and is present in blood and mucosal tissue, such as the lining of the lungs where they interfere with colonization of bacteria. The system can be initiated through three different pathways: the classical (CP), the lectin (LP) and the alternative pathway (AP). All pathways lead to a cascade of several protein-protein interactions and proteolytic steps leading to cleavage of C3 and C5 in biologically active components. C3a and C5a are able to attract inflammatory cells to the site of infection and activate them. C3b is another cleavage product of C3 and when deposited on the microbial surface stimulates bacterial uptake by phagocytes in a mechanism called opsonization. Besides opsonization and recruitment of phagocytes the complement system can also directly kill bacteria by forming pores into the microbial membrane and inducing lysis. The pore-forming heteromer is called the membrane attack complex (MAC) and is composed of C5b, C6, C7, C8 and C9. The classical pathway gets activated with binding of antibodies to the microbial surface. IgM or IgG bind to the first complement component C1q, which results in activation of serine proteases that cleave C2 and C4 which in turn form C4b2a, a C3 convertase. The same C3 convertase is formed upon activation through the lectin pathway, in which mannose-binding lectin-associated serine proteases cleave C4 and C2 upon binding of lectin to mannose on the bacterial surface. No specific signal is needed for the alternative pathway to form another C3 convertase: C3bBb, comprising of spontaneously hydrolyzed C3, or C3b generated by the CP/LP, and factor B that is activated by factor D. The C3 convertases (C4b2a and C3bBb) switch to C5 convertases in response to high levels of C3b deposition on the bacterial surface (Jongerius et al. 2007). The complement system does not only directly or indirectly kill bacteria but also has a role in regulating adaptive immunity. Besides regulating B-cell immunity, complement-mediated signaling directly stimulates and modulates T-cell responses (Jongerius et al. 2014) and promotes antigen processing by antigen-presenting cells (Serruto et al. 2010), indirectly influences T-cell activation. On the downside, over-activation or lack of down-regulation of the complement system can result in systemic inflammation during sepsis (Jongerius et al. 2007) and autoimmune diseases when host cells are recognized as non-self and damaged by complement activation (Meri et al. 2013).
Complement evasion by Bordetella pertussis
In order to survive and successfully colonize the host the bacteria must have strategies to evade the early immune response. It seems especially necessary for B. pertussis to prevent complement activation because it does not express antigen O (Marr et al. 2011). Antigen O is the chain of repetitive oligosaccharides of the lipopolysaccharides in the membrane of most Gram-negative bacteria and works as a protective shield. The Bordetella strains B. parapertussis and B. bronchoseptica express liposaccharide containing this O-antigen, but despite being more virulent B. pertussis lacks O-antigen in its cell wall. B. pertussis expresses a different kind of surface polysaccharide called Bps (Bordetella pertussis polysaccharide) which seems to provide serum resistance, since Bps mutant strains were more sensitive to complement-mediated killing then wild type bacteria (Ganguly et al. 2014). In addition to polysaccharides B. pertussis is known to express several other complement evasion molecules. Mutants lacking Bordetella autotransporter protein-C (BapC) were less resistant to serum killing, although the mechanism of resistance is yet to be identified (Noofeli et al. 2011). Another autotransporter of B. pertussis, Bordetella resistance to killing A (BrkA) that promotes attachment of the bacteria to human cells, is involved in complement evasion as well (Jongerius et al. 2014). Studies have shown that BrkA reduces C3 and C4 deposition and inhibits formation of MAC (Barnes et al. 2001). BrkA did not affect C1 deposition levels, suggesting BrkA only inhibits the classical pathway however the precise mechanism remains unknown (Barnes et al. 2001). Besides binding to complement components and thereby inhibiting their action by proteolysis or inducing conformational changes, bacteria can also employ a different strategy: to recruit and activate human complement inhibitors. The main regulators of the classical pathway are C4b-binding protein (C4BP) that B. pertussis binds to its surface with filamentous hemagglutinin (FHA) (Berggard et al. 2001), and C1 esterase inhibitor (C1-inh) recruited by B. pertussis autotransporter Vag8 (Marr et al. 2011). Factor H (fH) regulates the activity of the alternative pathway. The AP does not need a specific signal, making every unprotected surface vulnerable; therefore bacteria need to evade this pathway. Many bacteria do so by binding fH via a common binding site, forming a tripartite complex consisting of the microbial receptor, C3b and fH (Meri et al. 2013). This complex inhibits the opsonizing actions of C3b (Meri et al. 2013) and thereby protects the bacteria against phagocytosis. Recent studies show that B. pertussis binds fH at domain 20 (the common binding site) while fH remains its inhibiting properties (Amdahl et al. 2010), but a particular B. pertussis protein that binds fH has not been identified yet.
The whole cell pertussis (wP) vaccine was introduced in the 1950, efficiently lowering the cases of pertussis, which was the leading cause of child death before introduction of the vaccine. Unfortunately the vaccine was causing a lot of unwanted side effects in children. This had lead to introduction of an acellular pertussis (aP) vaccine, comprising of just several antigens. Although this vaccine provides a sufficient level of protection for infants to severe pertussis, the incidence of pertussis in adults has been increasing since the introduction of the aP vaccine in the 1990s (Mooi et al. 2013). More awareness and better diagnostics contribute to this increase in pertussis cases, but moreover the protection given by the aP is short-lived. Immunity generated by the wP vaccine is mediated largely by T-helper 1 cells (Th1), whereas aP vaccination results in antibodies that induce Th2 and Th17 responses. Natural immunity is associated with Th1 cells, and hence IFN-ï¿½ï¿½ production, as is seen in children recovering from whooping cough (Higgs et al. 2012). The adaptive immunity provided by aP vaccination is mainly humoral, promoted by Th2 cells that seem to be less efficient than Th1 cells in clearing B. pertussis infections (Higgs et al. 2012). In addition the provided immunity by aP vaccination is not lasting as long as the wP (Mooi et al. 2014). This occurrence of waning immunity is proposed to be the main cause of re-emergence of infections with B. pertussis. In addition, there is evidence the bacteria have been genetically adapting to the aP vaccines, caused by antigenic variation and selection pressure. In The Netherlands, France, Finland and Japan there are strains circulating that do not express FHA, PT and PRN, components of the aP vaccine used in these countries (Mooi et al. 2014). Future vaccines should be improved by, for example, adding other antigen preparations to provide long-lasting immunity, and additionally to strengthen the immune response. By strengthening innate immunity, the immune system will be able to directly eliminate the bacteria, preventing transmission and therefore also strain adaptation. The use of complement evasion molecules as vaccine targets would allow faster complement activation upon infection with B. pertussis. In this study we look to identify and characterize such attractive vaccine component candidates.
Prior to this characterization study a secretome phage display was constructed (Fevre et al. 2014). This phage display was used to identify potential immune evasion molecules. Phage display is the technique of expressing proteins on the surface of a bacteriophage. A phage library comprises of several phage clones that express different proteins, which retain their original shape and behavior. This allows selection of the displayed proteins based on affinity, with successive rounds of selection to identify proteins with high affinity for the selected targets. Secretome phage display, based on whole-genome phage display, is tailored for identification of immune evasion molecules with only secretome proteins being displayed. The bacterial secretome is the collection of all secreted and surface-bound proteins (Fevre et al. 2014). Phage display was performed for the secretome of Bordetella pertussis, with 3 rounds of selection and 7×107 phage clones. The phages were targeted on six components of the complement system: alternative pathway regulating factor H (fH), the opsonin C3b and C3 together with C5 (the two mixed) and C8 together with C9. After the third round there were five hits: BP0069, BP3355, BP0173, BP1251 and fhaC. Supposedly BP1251 and fhaC seem to bind C8 and/or C9, BP0069 binds factor H and BP0173 and BP3355 are suggested to bind a mix of complement components.
Analyzing information on chemical parameters (see table 1), gene alignment and genomic context (provided by NCBI Gene, NCBI Protein, BLAST and ExPASy ProtParam) might provide some clues of the functional properties of the proteins, but their complete function is yet unknown. BP0069 is a 43 kD protein which is 97% identical to ABC transporters from B. bronchispetica and around 30% identical to TRAP transporters from several bacteria species, both are transmembrane transporters for nutrients intake. BP3355 is a 18 kDa hypothetical protein, that has no evident homologues. BP0173 is a 24 kD hypothetical protein and BP1251 is a 26 kD putative toxin.