The spirochete Borrelia is the bacterial agent causing both Lyme disease (LD) and relapsing fever (RF) (Steere et al., 2004; Radolf et al., 2012). LD, the most common vector-borne illness in the U.S. and Europe, is caused by the Borrelia burgdorferi sensu lato complex, consisting of 20 species of which 6 cause illness in humans (Rudenko et al., 2011). B. burgdorferi sensu stricto (B. burgdorferi) causes most infections in the U.S., whereas this species as well as B. garinii and B. afzelii cause most infections in Europe (Baranton et al., 1992; Canica et al., 1993; Steere et al., 2004). LD borreliae are transmitted by Ixodes ticks to reservoir animals and humans (Steere et al., 2004). After a tick bite, the bacteria infect the skin at the feeding site, often accompanied with the development of an erythema migrans skin rash (Steere et al., 2004). If left untreated, LD borreliae are capable of disseminating to tissues and organs to cause diverse manifestations including arthritis, carditis, and neuroborreliosis (Steere et al., 2004). Human RF infections are transmitted by ticks or lice, resulting in tick-borne relapsing fever (TBRF), or louse-borne relapsing fever (LBRF; Cutler, 2015). At least 10 species of TBRF borreliae, including Borrelia hermsii, Borrelia parkeri, and B. duttonii, are transmitted through bites by various Ornithodoros ticks whereas LBRF B. recurrentis is solely transmitted by the clothing louse P. humanus via crushed lice or feces contacting irritated human skin. Upon transmission, RF borreliae cause bacteremia, and alternating febrile/afebrile episodes corresponding with antigenic variation (Cutler, 2015). The spirochetes then disseminate to the central nervous system and may lead to complications in the brain, lungs, kidneys, and spleen (Dworkin et al., 2008; Cutler, 2015).

Survival in the bloodstream is thought to be essential for LD and RF borreliae to cause systemic disease. The complement system is an innate immune defense mechanism in the bloodstream of humans and other vertebrate animals against pathogens (Zipfel and Skerka, 2009). The complement system can be activated via three pathways: classical, lectin, and alternative, all of which result in the formation of C3 convertases (Figure 1). The classical pathway is initiated by the active form of C1 complex (C1qr2s2) binding to antibody-bacterial antigen complexes. The lectin pathway is initiated by binding of lectins [mannan-binding lectin (MBL) or ficolins] to an MBL serine protease (MASP) and microbial carbohydrate. Activation of these pathways leads to the generation of the C3 convertase C4b2a. The alternative pathway is initiated by interaction of C3b with the microbial surface and generates the C3 convertase C3bBb. Both C3 convertases recruit C3b to form C5 convertases, which further promotes the formation of C5b-9 membrane attack complex (MAC) and pathogen lysis. The activation of complement also promotes the release of proinflammatory peptides (C3a and C5a) and deposition of opsonic C3b molecules on the microbial surface to enhance phagocytic clearance (Figure 1). To avoid potential self-damage due to complement activation, vertebrate animals produce a number of diverse complement regulators to negatively regulate the complement system (Figure 1). Examples include C1 inhibitor (C1-INH), which binds to inactive C1rs and/or MASP to block the initiation of the classical and/or lectin pathways. Factor H (FH) and FHL-1 (a truncated form of FH) both bind to and promote the cleavage of C3b via recruiting the protease factor I (FI) to prevent the formation of C3 convertase C3bBb. C4b-binding protein (C4BP) binds to and triggers the degradation of C4b via recruiting FI to inhibit the formation of the C3 convertase C4b2a. Lastly, CD59 binds to C8 and C9 to block the formation of the MAC to avoid lysis of host cells.

Bacterial pathogens, including LD borreliae, produce outer surface proteins that bind and recruit complement regulators on the cell surface to inhibit complement activation and prevent killing (Table 1 for references; Kraiczy, 2016). B. burgdorferi and B. garinii produce the C4BP-binding protein p43, which may recruit C4BP to the bacterial surface to promote C4b degradation and eventually inhibit both classical and lectin pathways. Except B. bavariensis, all other serum-resistant LD borreliae produce up to five Complement Regulator-Acquiring Surface Proteins (CRASPs): CRASP-1 (CspA), CRASP-2 (CspZ), CRASP-3 (ErpP), CRASP-4 (ErpC), and CRASP-5 (ErpA). CspA and CspZ bind FH (and/or FHL-1). These proteins simultaneously bind C3b and then promote C3b degradation on spirochete surface to downregulate the alternative pathway (Meri et al., 2013). ErpP, ErpC, and ErpA facilitate serum resistance of LD borreliae and bind to FH, but the biological significance of these interactions is unclear.

Similar to LD borreliae, RF borreliae produce complement regulator-binding proteins on their surface [Table 1 for references; Embers and SpringerLink, (Online service), 2012]. BHA007 in B. hermsii and its homolog CihC in both B. recurrentis and B. duttonii bind C4BP. CihC also binds C1-INH. The association of these proteins with C1-INH and C4BP on the surface of spirochetes prevents the formation of C1 and MASP complexes and induces the cleavage of C4b, respectively, to presumably inhibit the classical and lectin pathways. BhCRASP-1 and FhbA in B. hermsii, BpcA in B. parkeri, and HcpA in B. recurrentis bind FH (and/or FHL-1), which promotes C3b cleavage on bacterial surface and inhibits the alternative pathway.

LD borreliae also produce other outer surface proteins that interact with complement components to inhibit the formation of complement complexes and negatively modulate the complement system (Table 1 for references; Kraiczy, 2016). BBK32 of B. burgdorferi, known for both fibronectin (Probert and Johnson, 1998) and glycosaminoglycan binding (Fischer et al., 2006), was recently reported as a C1r-binding protein. By binding to the inactive form of C1r, BBK32 blocks the formation of the active C1 complex and inhibits the classical pathway. CspA of B. burgdorferi, B. afzelii, and B. spielmanii, and BGA66 and BGA71 of B. bavariensis, bind C7, C8, and C9. An unknown CD59-like protein of B. burgdorferi binds C9. These interactions result in the inhibition of MAC, thereby preventing bacteriolysis.

In the natural transmission of LD or RF borreliae from ticks to vertebrate animals, the spirochetes first colonize the skin at the tick feeding site prior to disseminating into the bloodstream and migrating into the surrounding tissues (Radolf et al., 2012; Coburn et al., 2013). In traditional models, mice are inoculated subcutaneously or intradermally, or by bite from a tick infected with LD or RF borreliae to study the contribution of spirochete factors during infection (Barthold et al., 1990; Simon et al., 1991). However, since a failure at either initial skin colonization or bloodstream survival would lead to low or undetectable bacterial burdens in the animal, it can be difficult to distinguish the roles of spirochete factors in promoting survival within the mammalian host using traditional models.

A short-term murine model has recently been developed using the LD spirochete B. burgdorferi to investigate the roles of bacterial outer surface proteins in mammalian bloodstream survival (Caine and Coburn, 2015). This model intravenously inoculates mice with a high number of spirochetes for up to 1 h. The ability of the spirochetes to survive in the bloodstream can be detected by measuring bacteremia (Caine and Coburn, 2015). Intravenous inoculation bypasses the initial step of skin colonization allowing the study of non-infectious mutant spirochetes. Therefore, this strategy teases apart the contributions of Borrelia factors with multiple functions in bloodstream survival, protein adhesion, and tissue attachment. For example, B. burgdorferi outer surface protein BBK32 contributes to colonization of the inoculation site of skin (Seshu et al., 2006; Hyde et al., 2011; Lin et al., 2015). Whether this protein contributes to mammalian bloodstream survival during Lyme infection is difficult to assess by subcutaneous needle or tick infection. Using short-term intravenous inoculation in a murine model, BBK32 ectopically-produced on a non-infectious, serum-sensitive B. burgdorferi strain promotes spirochete survival in the bloodstream (Caine and Coburn, 2015). This strategy has also been applied to identify the contribution of other B. burgdorferi factors in promoting bloodstream survival (Caine and Coburn, 2015). As RF borreliae are also blood-borne pathogens that disseminate into host tissues, this short-term model could be employed to further characterize serum resistance and disease progression in RF borreliae. Though some complement components or regulator are polymorphic among vertebrate animals (Lu et al., 2008), which raises a concern that the in vivo murine models may not be relevant to humans, recent developed humanized mouse strains may be utilized as a solution of this issue (Beernink et al., 2012).

Bloodstream survival of LD or RF borreliae is thought to be essential for spirochetes to survive in humans, and ultimately cause LD or RF disease manifestations. Serum resistance, adsorption, hemolytic, cofactor, and deposition assays, as well as a recently established short term intravenous inoculation murine model are all used to elucidate the mechanism of LD and RF borreliae evasion of the complement system and survival in the bloodstream. The data reviewed here are mainly on borrelial interactions with humans, but these assays can also be applied to the interactions with other vertebrate hosts, which will elucidate the role of the borrelial complement evasion in the enzootic cycle. Understanding these mechanisms in both humans and other vertebrate hosts will aid in the development of novel therapeutic strategies and disease prevention by targeting these complement component- or regulator-binding proteins to ultimately improve human health.

AM, PK, and YL wrote the manuscript; and AM and YL prepared the figure and table.